CN114638977A - Thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment - Google Patents
Thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment Download PDFInfo
- Publication number
- CN114638977A CN114638977A CN202210240410.8A CN202210240410A CN114638977A CN 114638977 A CN114638977 A CN 114638977A CN 202210240410 A CN202210240410 A CN 202210240410A CN 114638977 A CN114638977 A CN 114638977A
- Authority
- CN
- China
- Prior art keywords
- analyzed
- image
- target sample
- layer chromatography
- sample
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/22—Matching criteria, e.g. proximity measures
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Artificial Intelligence (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Evolutionary Computation (AREA)
- Bioinformatics & Computational Biology (AREA)
- Computational Linguistics (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Health & Medical Sciences (AREA)
- Biomedical Technology (AREA)
- Biophysics (AREA)
- Evolutionary Biology (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment. The thin-layer chromatography component analysis method based on the adaptive weight fusion carries out feature vector extraction and spectral band extraction on the thin-layer chromatography image of a sample to be analyzed and the thin-layer chromatography image of a target sample with known components, uses the feature vector matching and the relative position relationship of spectral bands as identification factors to input the identification factors into a neural network for component identification, fuses the feature point similarity and the spectral band position relationship similarity, and can improve the accuracy and efficiency of the thin-layer chromatography component analysis result.
Description
Technical Field
The invention relates to the technical field of chemical component analysis, in particular to a thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment.
Background
The thin-layer chromatography component analysis refers to a technology of coating a stationary phase on an experimental plate, using a corresponding solvent as a mobile phase to form a uniform thin layer to analyze components of a sample, comparing a chromatogram of the sample to be detected with a chromatogram of a reference substance obtained by the same method, and making a differential identification and analysis by using the same spots on the chromatograms of a real sample and the reference substance at corresponding positions, thereby being widely applied to the field of component analysis of medicinal materials.
In the existing thin-layer chromatography component analysis, manual comparison is often adopted, but because the imaging condition of a chromatographic band of the thin-layer chromatography is greatly influenced by the outside world, the chromatographic band often has the characteristics of blotch, irregularity and large strength difference, so that the manual identification difficulty is high and the efficiency is low. Moreover, manual judgment has certain subjectivity due to the influences of various factors such as distortion, illumination and environmental humidity of actual chromatograms, and according to different knowledge reserves and identification experiences of different firmed people, the same chromatogram can generate various results, so that the accuracy and stability of manual comparison are difficult to effectively guarantee.
Thus, there is a need for improvements and enhancements in the art.
Disclosure of Invention
Aiming at the defects in the prior art, the invention provides a thin-layer chromatography component analysis method based on adaptive weight fusion, and aims to solve the problem of low efficiency of manual thin-layer chromatography component analysis in the prior art.
In order to solve the technical problems, the technical scheme adopted by the invention is as follows:
in a first aspect of the present invention, there is provided a thin layer chromatography component analysis method based on adaptive weight fusion, the method comprising:
acquiring an image to be analyzed, wherein the image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, inputting the image to be analyzed into a preset filter, and extracting each first feature vector in the image to be analyzed through the preset filter;
matching the first characteristic vector with each template characteristic vector of a target sample, and acquiring a first similarity score of the sample to be analyzed and the target sample based on a matching result, wherein the template characteristic vector is extracted from a thin-layer chromatography image of the target sample;
inputting the image to be analyzed into a trained first neural network, and acquiring each first spectral band in the image to be analyzed output by the first neural network;
obtaining a second similarity score of the sample to be analyzed and the target sample based on the relative position between the first spectral bands and the relative position between template spectral bands of the target sample, wherein the template spectral bands of the target sample are spectral bands extracted from a thin-layer chromatography image of the target sample;
and inputting the first similarity score and the second similarity score of the sample to be analyzed and each target sample into a trained second neural network, and obtaining a component analysis result output by the second neural network.
The thin layer chromatography component analysis method based on adaptive weight fusion, wherein before the matching of the first feature vector with each template feature vector of a target sample, the method further comprises:
inputting the thin-layer chromatography image of the target sample into the preset filter, and extracting each second feature vector in the thin-layer chromatography image of the target sample through the preset filter;
and selecting at least one second feature vector as the template feature vector corresponding to the target sample according to the modulus of each second feature vector.
The thin layer chromatography component analysis method based on adaptive weight fusion, wherein the matching of the first feature vector with each template feature vector of a target sample, comprises:
calculating the similarity between the first feature vector and each template feature vector;
when the similarity between the first feature vector and the template feature vector is larger than a preset threshold value, determining that the first feature vector is matched with the template feature vector.
The thin layer chromatography component analysis method based on adaptive weight fusion, wherein the obtaining of the first similarity score of the sample to be analyzed and the target sample based on the matching result comprises:
and obtaining the first similarity score according to the number of the first feature vectors matched with the template feature vectors of the target sample.
The thin layer chromatography component analysis method based on adaptive weight fusion, wherein the obtaining of the second similarity score of the sample to be analyzed and the target sample based on the relative position between the first spectral bands and the relative position between the template spectral bands of the target sample comprises:
calculating the difference of position coordinates between the adjacent first spectral bands to obtain first relative position data among the first spectral bands;
calculating the difference of position coordinates between adjacent template spectral bands to obtain second relative position data between the template spectral bands;
and acquiring the second similarity score according to the first relative position data and the second relative position data.
The thin layer chromatography component analysis method based on adaptive weight fusion, wherein before obtaining the second similarity scores of the sample to be analyzed and the target sample based on the relative positions between the first spectral bands and the relative positions between the template spectral bands, the method further comprises:
inputting the thin-layer chromatography image of the target sample into the trained first neural network, and acquiring each template spectral band of the target sample output by the first neural network.
The thin-layer chromatography component analysis method based on the adaptive weight fusion is characterized in that parameters of the preset filter and parameters of the first neural network are obtained through training, training data comprise multiple groups of sample data, each group of sample data comprises an image to be analyzed, and each target sample and components of the sample corresponding to the image to be analyzed.
In a second aspect of the present invention, there is provided a thin-layer chromatography component analysis apparatus based on adaptive weight fusion, comprising:
the device comprises a feature extraction module, a comparison module and a comparison module, wherein the feature extraction module is used for acquiring an image to be analyzed, the image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, the image to be analyzed is input to a preset filter, and each first feature vector in the image to be analyzed is extracted through the preset filter;
a first similarity module, configured to match the first feature vector with each template feature vector of a target sample, and obtain a first similarity score between the sample to be analyzed and the target sample based on a matching result, where the template feature vector is a feature vector extracted from a thin layer chromatography image of the target sample;
the spectral band extraction module is used for inputting the image to be analyzed into a trained first neural network and acquiring each first spectral band in the image to be analyzed output by the first neural network;
a second similarity module, configured to obtain second similarity scores of the sample to be analyzed and the target sample based on a relative position between each of the first spectral bands and a relative position between each of template spectral bands of the target sample, where the template spectral bands of the target sample are spectral bands extracted from a thin-layer chromatography image of the target sample;
and the fusion module is used for inputting the first similarity scores and the second similarity scores of the samples to be analyzed and the target samples into a trained second neural network and acquiring component analysis results output by the second neural network.
In a third aspect of the present invention, there is provided a terminal, including a processor, and a computer-readable storage medium communicatively connected to the processor, the computer-readable storage medium being adapted to store a plurality of instructions, and the processor being adapted to call the instructions in the computer-readable storage medium to execute the steps of implementing any one of the above-mentioned thin layer chromatography component analysis methods based on adaptive weight fusion.
In a fourth aspect of the present invention, there is provided a computer readable storage medium storing one or more programs, the one or more programs being executable by one or more processors to implement the steps of the method for thin layer chromatography component analysis based on adaptive weight fusion as described in any one of the above.
Compared with the prior art, the invention provides a thin-layer chromatographic component analysis method based on adaptive weight fusion and related equipment, wherein in the thin-layer chromatographic component analysis method based on adaptive weight fusion, a thin-layer chromatographic image of a sample to be analyzed and a thin-layer chromatographic image of a target sample of known components are subjected to feature vector extraction and spectral band extraction, and feature vector matching and the relative position relationship of spectral bands are used as identification elements and input into a neural network for component identification, so that the feature point similarity and the spectral band position relationship similarity are fused, and the accuracy and the efficiency of a thin-layer chromatographic component analysis result can be improved.
Drawings
FIG. 1 is a flow chart of an embodiment of a thin layer chromatography component analysis method based on adaptive weight fusion provided by the present invention;
FIG. 2 is a schematic diagram of feature vector matching in an embodiment of a thin-layer chromatography component analysis method based on adaptive weight fusion according to the present invention;
FIG. 3 is a schematic diagram of a second neural network in an embodiment of the thin layer chromatography component analysis method based on adaptive weight fusion provided by the present invention;
FIG. 4 is a schematic structural diagram of an embodiment of a thin-layer chromatography component analysis device based on adaptive weight fusion provided by the invention;
fig. 5 is a schematic diagram of an embodiment of a terminal provided in the present invention.
Detailed Description
In order to make the objects, technical solutions and effects of the present invention clearer and clearer, the present invention is further described in detail below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
The thin-layer chromatographic component analysis method based on adaptive weight fusion provided by the invention can be applied to a terminal with computing power, the terminal can execute the thin-layer chromatographic component analysis method based on adaptive weight fusion provided by the invention to generate a component analysis result of a sample to be analyzed, and the terminal can be but is not limited to various computers, mobile terminals, intelligent household appliances, wearable equipment and the like.
Example one
As shown in fig. 1, in an embodiment of the thin layer chromatography component analysis method based on adaptive weight fusion, the method comprises the following steps:
s100, obtaining an image to be analyzed, inputting the image to be analyzed into a preset filter, and extracting each first feature vector in the image to be analyzed through the preset filter.
The image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, and in order to analyze components of the sample to be analyzed, the thin-layer chromatography image is firstly manufactured according to the sample to be analyzed, so that the image to be analyzed is obtained.
In this embodiment, first, feature point extraction is performed on the image to be analyzed to obtain a plurality of feature vectors, specifically, the image to be analyzed is input to a preset filter, each feature point in the image to be analyzed is extracted through the preset filter, a feature value of each feature point is a feature vector, the preset filter may use an existing filter, in order to enable the preset filter to extract a more accurate feature vector, in this embodiment, parameters of the preset filter are obtained through training, the preset filter may be trained alone, when training alone, training data used for training the preset filter may include a sample image and a labeled feature corresponding to the sample image, and the preset filter may also be trained together with other neural networks used in the method, details will be described later.
S200, matching the first feature vector with each template feature vector of a target sample, and acquiring a first similarity score of the sample to be analyzed and the target sample based on a matching result.
The template feature vector is a feature vector extracted from a thin layer chromatography image of the target sample, the target sample is a sample with known components, the target sample can be multiple, and each target sample corresponds to multiple template feature vectors. Specifically, the thin-layer chromatography image of the target sample is also input to the preset filter, and the template feature vector corresponding to the thin-layer chromatography image of the target sample is extracted through the preset filter, that is, before the first feature vector is matched with the template feature vector, the thin-layer chromatography component analysis method based on adaptive weight fusion provided in this embodiment further includes the steps of:
inputting the thin-layer chromatography image of the target sample into the preset filter, and extracting each second feature vector in the thin-layer chromatography image of the target sample through the preset filter;
and selecting at least one second feature vector as the template feature vector corresponding to the target sample according to the modulus of each second feature vector.
That is to say, in this embodiment, the twin parallel filter is adopted to simultaneously extract the features of the thin-layer chromatography images of the sample to be analyzed and the target sample, so that the efficiency is high, and the extracted features can better extract the discrimination information of the sample and the template, thereby improving the matching accuracy.
Since there may be problems of tilt, distortion, bright-area misalignment, etc. in the chromatographic image, in this embodiment, before the image to be analyzed and the thin-layer chromatographic image of the target sample are input to the preset filter, the image to be analyzed and the thin-layer chromatographic image of the target sample are also preprocessed. Specific preprocessing operations include, but are not limited to, complete correction, length cutting and the like, and the problems of inclination, distortion, bright area dislocation and the like of the chromatogram are solved as much as possible through preprocessing. After pre-processing the image, the image is also scaled to a preset size, that is, such that the size of the image to be analyzed input to the preset filter and the thin layer chromatography image of the target sample are the same.
After obtaining each second feature vector corresponding to the target sample, each second feature vector may be directly used as one template feature vector, but a plurality of feature points and corresponding feature vectors may be extracted from one image through the preset filter. Specifically, a modulus of each second feature vector corresponding to the target sample is obtained, K second feature vectors with the largest modulus are selected as the target feature vectors corresponding to the target sample, and K is greater than or equal to 1. In the embodiment, the significant point measurement method is adopted to pay more attention to the important area in the thin-layer chromatogram, so that the similarity between the sample and the template can be more accurately reflected, and the recall rate is improved.
As shown in fig. 2, after obtaining each of the first feature vectors and the template feature vector of the target sample, matching the first feature vectors and the template feature vectors. Specifically, the matching the first feature vector with each template feature vector of the target sample includes:
calculating the similarity between the first feature vector and each template feature vector;
when the similarity between the first feature vector and the target feature vector is larger than a preset threshold value, determining that the first feature vector is matched with the target feature vector.
Calculating the similarity between the first feature vector and each template feature vector may be implemented by using a cosine similarity calculation method, or may be implemented by using other similarity calculation methods, for each first feature vector, calculating the similarity with each template feature vector of the target sample, and if the similarity between one first feature vector and one template feature vector is greater than a preset threshold, determining that the two match. Further, if the similarity between one of the first feature vectors and the plurality of template feature vectors is greater than the preset threshold, selecting the template feature vector with the greater similarity as the template feature vector matched with the first feature vector, that is, determining that the first feature vector is matched with the template feature vector with the greater similarity and is not matched with the rest of the template feature vectors.
Obtaining the first similarity score of the sample to be analyzed and the target sample according to the matching structure of the first feature vector and each target feature vector of the target sample, specifically including:
and obtaining the first similarity score according to the number of the first feature vectors matched with the template feature vectors of the target sample.
That is, the greater the number of the first feature vectors that can be successfully matched with the template feature vector of the target sample, the higher the first similarity score.
Referring to fig. 1 again, the thin layer chromatography component analysis method based on adaptive weight fusion provided in this embodiment further includes the steps of:
s300, inputting the image to be analyzed into the trained first neural network, and acquiring each first spectral band in the image to be analyzed output by the first neural network.
In this embodiment, in addition to determining the similarity between the sample to be analyzed and the target sample by using a feature point matching method, the similarity between the sample to be analyzed and the target sample is also determined by combining spectral band distribution characteristics in a chromatogram. Specifically, in the present embodiment, the first neural network is used to extract key spectral bands in the chromatogram. Specifically, the first neural network is trained by using a plurality of groups of first training data in advance, each group of the first training data includes a sample chromatogram and a labeled spectral band corresponding to the sample chromatogram, and the first training data can be obtained by manually labeling a key spectral band in the sample chromatogram. The structure of the first neural network may adopt the structure of an existing neural network, for example, the first neural network adopts a yolov3 detection network. Inputting the image to be analyzed to the trained first neural network, wherein the first neural network can extract a key spectral band in the image to be analyzed, and the key spectral band in the image to be analyzed extracted by the first neural network is called a first spectral band.
Referring to fig. 1 again, the method provided in this embodiment further includes the steps of:
s400, obtaining second similarity scores of the sample to be analyzed and the target sample based on the relative position between the first spectral bands and the relative position between the template spectral bands of the target sample.
The template spectral band of the target sample is a spectral band extracted from a thin-layer chromatographic image of the target sample, the thin-layer chromatographic image of the target sample is input into the first neural network, the first neural network can extract a key spectral band in the thin-layer chromatographic image of the target sample, and the key spectral band in the thin-layer chromatographic image of the target sample extracted by the first neural network is called as the template spectral band of the target sample. That is, before the obtaining of the second similarity score between the sample to be analyzed and the target sample based on the relative position between each of the first spectral bands and the relative position between each of the template spectral bands, the method further includes:
inputting the thin-layer chromatography image of the target sample into the trained first neural network, and acquiring each template spectral band of the target sample output by the first neural network.
The obtaining of the second similarity scores of the sample to be analyzed and the target sample based on the relative positions between the respective first spectral bands and the relative positions between the respective template spectral bands of the target sample comprises:
calculating the difference of position coordinates between the adjacent first spectral bands to obtain first relative position data among the first spectral bands;
calculating the difference of position coordinates between adjacent template spectral bands to obtain second relative position data between the template spectral bands;
and acquiring the second similarity score according to the first relative position data and the second relative position data.
After the key spectral band in the image to be analyzed and the key spectral band in the thin-layer chromatography image of the target sample are obtained, the difference between the position coordinates of the adjacent first spectral bands in the image to be analyzed is calculated, that is, the position coordinates of the adjacent first spectral bands are subtracted, so that a vector representing the relative position relationship between the first spectral bands can be obtained, and the vector is used as the first relative position data. In the same manner, the second relative position data can be obtained. And calculating the similarity between the first relative position data and the second relative position data to obtain the second similarity score, wherein the second similarity score can be calculated by a cosine similarity calculation method or other similarity calculation methods.
The relative position relation of the chromatographic bands is used as a key identification factor, so that the influence of the deformation of the chromatographic bands on the identification precision in actual sampling can be relieved.
Through the steps, the first similarity score and the second similarity score of the sample to be analyzed and the target sample can be obtained through calculation, and the first similarity score and the second similarity score of the sample to be analyzed and each target sample are obtained through calculation by the same method. The method provided by the embodiment further comprises the following steps:
s400, inputting the first similarity scores and the second similarity scores of the samples to be analyzed and the target samples into a trained second neural network, and obtaining component analysis results output by the second neural network.
And inputting the first similarity score and the second similarity score of the sample to be analyzed and each target sample into the second neural network, and obtaining a component analysis result of the sample to be analyzed through the second neural network.
The parameters of the preset filter and the parameters of the first neural network are obtained through training, the training data comprise a plurality of groups of sample data, each group of sample data comprises an image to be analyzed of a sample, and each target sample and the components of the sample corresponding to the image to be analyzed of the sample.
Specifically, the second neural network is obtained by training a plurality of groups of sample data, each group of sample data includes an image to be analyzed, and each target sample and a component of a sample corresponding to the image to be analyzed. Specifically, in the training process of the second neural network, the processing mode of steps S100 to S300 is adopted for the image to be analyzed of the sample, the first similarity score and the second similarity score corresponding to each target sample are obtained, and are input to the MLP network together with the class label, forward propagation is performed to obtain a prediction classification class, the size of a loss function between a predicted value and a true value is calculated by using a softmax loss function, and backward propagation is performed to update the network parameters. The parameters of the preset filter and the parameters of the second neural network can be updated together, and when iteration is performed for a certain turn, the training is stopped after the network parameters and the loss function tend to be stable. In manual identification, identification strategies can be divided into three categories, namely, identification of thin-layer chromatography by using a chromatographic band characteristic point preferentially, identification of a chromatographic band by using a chromatographic band relative position preferentially and identification of thin-layer chromatography by using the chromatographic band characteristic point and the chromatographic band relative position equally and with importance according to the importance degree of the chromatographic band characteristic point and the chromatographic band relative position relationship. During training, the relative position relation similarity scores and the feature point similarity scores of the spectral bands identified by different strategies and the training data of the corresponding chemical component types are collected. Therefore, the MLP network obtained through final training can perform adaptive weight fusion through forward propagation according to the similarity vector of the relative position relationship between the input chromatographic band characteristic points and the chromatographic bands, and an identification result fusing the advantages of two identification elements is obtained.
As shown in fig. 3, the second neural network may employ an MLP network structure. The second neural network can fuse the similarity of the characteristic points and the similarity of the spectral band position relationship, and can effectively improve the accuracy of the component analysis result.
In summary, the embodiment provides a thin-layer chromatography component analysis method based on adaptive weight fusion, which performs feature vector extraction and spectral band extraction on a thin-layer chromatography image of a sample to be analyzed and a thin-layer chromatography image of a target sample with known components, and uses the feature vector matching and the relative position relationship of spectral bands as identification elements to input the identification elements into a neural network for component identification, so that the feature point similarity and the spectral band position relationship similarity are fused, and the accuracy and efficiency of a thin-layer chromatography component analysis result can be improved.
It should be understood that, although the steps in the flowcharts shown in the figures of the present specification are shown in order as indicated by the arrows, the steps are not necessarily performed in order as indicated by the arrows. The steps are not performed in the exact order shown and described, and may be performed in other orders, unless explicitly stated otherwise. Moreover, at least a portion of the steps in the flowchart may include multiple sub-steps or multiple stages, which are not necessarily performed at the same time, but may be performed at different times, and the order of performing the sub-steps or stages is not necessarily sequential, but may be performed alternately or alternately with other steps or at least a portion of the sub-steps or stages of other steps.
It will be understood by those skilled in the art that all or part of the processes of the methods of the embodiments described above can be implemented by hardware instructions of a computer program, which can be stored in a non-volatile computer-readable storage medium, and when executed, can include the processes of the embodiments of the methods described above. Any reference to memory, storage, databases, or other media used in embodiments provided herein may include non-volatile and/or volatile memory. Non-volatile memory can include read-only memory (ROM), Programmable ROM (PROM), Electrically Programmable ROM (EPROM), Electrically Erasable Programmable ROM (EEPROM), or flash memory. Volatile memory can include Random Access Memory (RAM) or external cache memory. By way of illustration and not limitation, RAM is available in a variety of forms such as Static RAM (SRAM), Dynamic RAM (DRAM), Synchronous DRAM (SDRAM), Double Data Rate SDRAM (DDRSDRAM), Enhanced SDRAM (ESDRAM), Synchronous Link DRAM (SLDRAM), Rambus Direct RAM (RDRAM), direct bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
Example two
Based on the foregoing embodiment, the present invention further provides a thin-layer chromatography component analysis apparatus based on adaptive weight fusion, as shown in fig. 4, the thin-layer chromatography component analysis apparatus based on adaptive weight fusion includes:
the characteristic extraction module is used for acquiring an image to be analyzed, wherein the image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, the image to be analyzed is input into a preset filter, and each first characteristic vector in the image to be analyzed is extracted through the preset filter, specifically as described in embodiment one;
a first similarity module, configured to match the first feature vector with each template feature vector of a target sample, and obtain a first similarity score between the sample to be analyzed and the target sample based on a matching result, where the template feature vector is a feature vector extracted from a thin layer chromatography image of the target sample, as described in embodiment one;
a spectral band extraction module, configured to input the image to be analyzed to a trained first neural network, and obtain each first spectral band in the image to be analyzed output by the first neural network, as described in embodiment one;
a second similarity module, configured to obtain second similarity scores of the sample to be analyzed and the target sample based on a relative position between each of the first spectral bands and a relative position between each of template spectral bands of the target sample, where the template spectral bands of the target sample are spectral bands extracted from a thin-layer chromatography image of the target sample, as described in embodiment one;
a fusion module, configured to input the first similarity score and the second similarity score of the sample to be analyzed and each target sample to a trained second neural network, and obtain a component analysis result output by the second neural network, which is specifically described in embodiment one.
EXAMPLE III
Based on the above embodiments, the present invention further provides a terminal, as shown in fig. 5, where the terminal includes a processor 10 and a memory 20. Fig. 5 shows only some of the components of the terminal, but it is to be understood that not all of the shown components are required to be implemented, and that more or fewer components may be implemented instead.
The memory 20 may in some embodiments be an internal storage unit of the terminal, such as a hard disk or a memory of the terminal. The memory 20 may also be an external storage device of the terminal in other embodiments, such as a plug-in hard disk, a Smart Media Card (SMC), a Secure Digital (SD) Card, a Flash memory Card (Flash Card), and the like provided on the terminal. Further, the memory 20 may also include both an internal storage unit and an external storage device of the terminal. The memory 20 is used for storing application software installed in the terminal and various data. The memory 20 may also be used to temporarily store data that has been output or is to be output. In one embodiment, the memory 20 stores an adaptive weight fusion based thin layer chromatography component analysis program 30, and the adaptive weight fusion based thin layer chromatography component analysis program 30 is executable by the processor 10 to implement the adaptive weight fusion based thin layer chromatography component analysis method of the present application.
The processor 10 may be, in some embodiments, a Central Processing Unit (CPU), microprocessor or other chip for running program code stored in the memory 20 or Processing data, such as executing the target-based multimodal named entity recognition method.
In one embodiment, the following steps are implemented when the processor 10 executes the adaptive weight fusion based thin layer chromatography component analysis program 30 in the memory 20:
acquiring an image to be analyzed, wherein the image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, inputting the image to be analyzed into a preset filter, and extracting each first feature vector in the image to be analyzed through the preset filter;
matching the first characteristic vector with each template characteristic vector of a target sample, and acquiring a first similarity score of the sample to be analyzed and the target sample based on a matching result, wherein the template characteristic vector is extracted from a thin-layer chromatography image of the target sample;
inputting the image to be analyzed into a trained first neural network, and acquiring each first spectral band in the image to be analyzed output by the first neural network;
obtaining a second similarity score of the sample to be analyzed and the target sample based on the relative position between the first spectral bands and the relative position between template spectral bands of the target sample, wherein the template spectral bands of the target sample are spectral bands extracted from a thin-layer chromatography image of the target sample;
and inputting the first similarity score and the second similarity score of the sample to be analyzed and each target sample into a trained second neural network, and obtaining a component analysis result output by the second neural network.
Wherein, before matching the first feature vector with each template feature vector of a target sample, the method further comprises:
inputting the thin-layer chromatography image of the target sample into the preset filter, and extracting each second feature vector in the thin-layer chromatography image of the target sample through the preset filter;
and selecting at least one second feature vector as the template feature vector corresponding to the target sample according to the modulus of each second feature vector.
Wherein the matching the first feature vector with each template feature vector of a target sample comprises:
calculating the similarity between the first feature vector and each template feature vector;
when the similarity between the first feature vector and the template feature vector is larger than a preset threshold value, determining that the first feature vector is matched with the template feature vector.
Wherein the obtaining a first similarity score of the sample to be analyzed and the target sample based on the matching result comprises:
and obtaining the first similarity score according to the number of the first feature vectors matched with the template feature vectors of the target sample.
Wherein the obtaining of the second similarity scores of the sample to be analyzed and the target sample based on the relative positions between the respective first spectral bands and the relative positions between the respective template spectral bands of the target sample comprises:
calculating the difference of position coordinates between the adjacent first spectral bands to obtain first relative position data among the first spectral bands;
calculating the difference of position coordinates between adjacent template spectral bands to obtain second relative position data between the template spectral bands;
and acquiring the second similarity score according to the first relative position data and the second relative position data.
Wherein before obtaining the second similarity score between the sample to be analyzed and the target sample based on the relative position between each of the first spectral bands and the relative position between each of the template spectral bands, the method further comprises:
inputting the thin-layer chromatography image of the target sample into the trained first neural network, and acquiring each template spectral band of the target sample output by the first neural network.
The parameters of the preset filter and the parameters of the first neural network are obtained through training, the training data comprise a plurality of groups of sample data, each group of sample data comprises an image to be analyzed, and each target sample and the components of the sample corresponding to the image to be analyzed.
Example four
The present invention also provides a computer readable storage medium having stored thereon one or more programs, which are executable by one or more processors, to implement the steps of the adaptive weight fusion based thin layer chromatography component analysis method as described above.
Finally, it should be noted that: the above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A thin-layer chromatography component analysis method based on adaptive weight fusion, which is characterized by comprising the following steps:
acquiring an image to be analyzed, wherein the image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, inputting the image to be analyzed into a preset filter, and extracting each first feature vector in the image to be analyzed through the preset filter;
matching the first characteristic vector with each template characteristic vector of a target sample, and acquiring a first similarity score of the sample to be analyzed and the target sample based on a matching result, wherein the template characteristic vector is extracted from a thin-layer chromatography image of the target sample;
inputting the image to be analyzed into a trained first neural network, and acquiring each first spectral band in the image to be analyzed output by the first neural network;
obtaining a second similarity score of the sample to be analyzed and the target sample based on the relative position between the first spectral bands and the relative position between template spectral bands of the target sample, wherein the template spectral bands of the target sample are spectral bands extracted from a thin-layer chromatography image of the target sample;
and inputting the first similarity score and the second similarity score of the sample to be analyzed and each target sample into a trained second neural network, and obtaining a component analysis result output by the second neural network.
2. The method for thin layer chromatography component analysis based on adaptive weight fusion of claim 1, wherein before the matching of the first feature vector with each template feature vector of a target sample, the method further comprises:
inputting the thin-layer chromatography image of the target sample into the preset filter, and extracting each second feature vector in the thin-layer chromatography image of the target sample through the preset filter;
and selecting at least one second feature vector as the template feature vector corresponding to the target sample according to the modulus of each second feature vector.
3. The method for thin layer chromatography component analysis based on adaptive weight fusion of claim 1, wherein the matching the first feature vector with each template feature vector of a target sample comprises:
calculating the similarity between the first feature vector and each template feature vector;
when the similarity between the first feature vector and the template feature vector is larger than a preset threshold value, determining that the first feature vector is matched with the template feature vector.
4. The thin layer chromatography component analysis method based on adaptive weight fusion of claim 1, wherein the obtaining of the first similarity score of the sample to be analyzed and the target sample based on the matching result comprises:
and obtaining the first similarity score according to the number of the first feature vectors matched with the template feature vectors of the target sample.
5. The adaptive weight fusion-based thin layer chromatography component analysis method of claim 1, wherein the obtaining of the second similarity scores for the sample to be analyzed and the target sample based on the relative positions between the respective first spectral bands and the relative positions between the respective template spectral bands of the target sample comprises:
calculating the difference of position coordinates between the adjacent first spectral bands to obtain first relative position data among the first spectral bands;
calculating the difference of position coordinates between adjacent template spectral bands to obtain second relative position data between the template spectral bands;
and acquiring the second similarity score according to the first relative position data and the second relative position data.
6. The method for thin layer chromatography component analysis based on adaptive weight fusion of claim 1, wherein before the obtaining the second similarity score between the sample to be analyzed and the target sample based on the relative position between the first spectral bands and the relative position between the template spectral bands, the method further comprises:
inputting the thin-layer chromatography image of the target sample into the trained first neural network, and acquiring each template spectral band of the target sample output by the first neural network.
7. The thin layer chromatography component analysis method based on adaptive weight fusion according to any one of claims 1-6, wherein the parameters of the preset filter and the parameters of the first neural network are obtained by training together, the training data comprises a plurality of groups of sample data, each group of sample data comprises an image to be analyzed, and each target sample and the component of the sample corresponding to the image to be analyzed.
8. A thin layer chromatography component analysis device based on adaptive weight fusion, comprising:
the device comprises a feature extraction module, a feature extraction module and a feature extraction module, wherein the feature extraction module is used for acquiring an image to be analyzed, the image to be analyzed is a thin-layer chromatography image of a sample to be analyzed, the image to be analyzed is input to a preset filter, and each first feature vector in the image to be analyzed is extracted through the preset filter;
a first similarity module, configured to match the first feature vector with each template feature vector of a target sample, and obtain a first similarity score between the sample to be analyzed and the target sample based on a matching result, where the template feature vector is a feature vector extracted from a thin-layer chromatography image of the target sample;
the spectral band extraction module is used for inputting the image to be analyzed into a trained first neural network and acquiring each first spectral band in the image to be analyzed output by the first neural network;
a second similarity module, configured to obtain second similarity scores of the sample to be analyzed and the target sample based on a relative position between each of the first spectral bands and a relative position between each of template spectral bands of the target sample, where the template spectral bands of the target sample are spectral bands extracted from a thin-layer chromatography image of the target sample;
and the fusion module is used for inputting the first similarity scores and the second similarity scores of the samples to be analyzed and the target samples into a trained second neural network and acquiring component analysis results output by the second neural network.
9. A terminal, characterized in that the terminal comprises: a processor, a computer readable storage medium communicatively connected to the processor, the computer readable storage medium adapted to store a plurality of instructions, the processor adapted to invoke the instructions in the computer readable storage medium to perform the steps of implementing the method for thin layer chromatography component analysis based on adaptive weight fusion of any of claims 1-7 above.
10. A computer readable storage medium storing one or more programs, the one or more programs being executable by one or more processors to perform the steps of the adaptive weight fusion based thin layer chromatography component analysis method of any one of claims 1-7.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210240410.8A CN114638977A (en) | 2022-03-10 | 2022-03-10 | Thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN202210240410.8A CN114638977A (en) | 2022-03-10 | 2022-03-10 | Thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment |
Publications (1)
Publication Number | Publication Date |
---|---|
CN114638977A true CN114638977A (en) | 2022-06-17 |
Family
ID=81947767
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN202210240410.8A Pending CN114638977A (en) | 2022-03-10 | 2022-03-10 | Thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN114638977A (en) |
-
2022
- 2022-03-10 CN CN202210240410.8A patent/CN114638977A/en active Pending
Similar Documents
Publication | Publication Date | Title |
---|---|---|
CN112560876B (en) | Single-stage small sample target detection method for decoupling measurement | |
CN109615016B (en) | Target detection method of convolutional neural network based on pyramid input gain | |
CN111696094B (en) | Immunohistochemical PD-L1 membrane staining pathological section image processing method, device and equipment | |
CN110674712A (en) | Interactive behavior recognition method and device, computer equipment and storage medium | |
Li et al. | An anti-fraud system for car insurance claim based on visual evidence | |
US11686703B2 (en) | Automated analysis of analytical gels and blots | |
CN110287936B (en) | Image detection method, device, equipment and storage medium | |
CN111832581B (en) | Lung feature recognition method and device, computer equipment and storage medium | |
CN113239975B (en) | Target detection method and device based on neural network | |
CN109241998B (en) | Model training method, device, equipment and storage medium | |
CN113989604A (en) | Tire DOT information identification method based on end-to-end deep learning | |
CN112668462A (en) | Vehicle loss detection model training method, vehicle loss detection device, vehicle loss detection equipment and vehicle loss detection medium | |
CN113158777A (en) | Quality scoring method, quality scoring model training method and related device | |
Valledor et al. | Standardization of data processing and statistical analysis in comparative plant proteomics experiment | |
Muzakir et al. | Model for Identification and Prediction of Leaf Patterns: Preliminary Study for Improvement | |
WO2024051427A1 (en) | Coin identification method and system, and storage medium | |
CN111444833B (en) | Fruit actual measurement production method, device, computer equipment and storage medium | |
CN114638977A (en) | Thin-layer chromatography component analysis method based on adaptive weight fusion and related equipment | |
Arora et al. | An algorithm to straighten the bent human chromosomes | |
CN116958604A (en) | Power transmission line image matching method, device, medium and equipment | |
CN116843677A (en) | Appearance quality detection system and method for sheet metal part | |
US10140727B2 (en) | Image target relative position determining method, device, and system thereof | |
Zou et al. | YOLOv7‐EAS: A Small Target Detection of Camera Module Surface Based on Improved YOLOv7 | |
CN112801013B (en) | Face recognition method, system and device based on key point recognition verification | |
CN114757941A (en) | Transformer substation equipment defect identification method and device, electronic equipment and storage medium |
Legal Events
Date | Code | Title | Description |
---|---|---|---|
PB01 | Publication | ||
PB01 | Publication | ||
SE01 | Entry into force of request for substantive examination | ||
SE01 | Entry into force of request for substantive examination |