CN115855869A - Detection device and method of gas-phase mass spectrometer for tree species - Google Patents
Detection device and method of gas-phase mass spectrometer for tree species Download PDFInfo
- Publication number
- CN115855869A CN115855869A CN202211447444.0A CN202211447444A CN115855869A CN 115855869 A CN115855869 A CN 115855869A CN 202211447444 A CN202211447444 A CN 202211447444A CN 115855869 A CN115855869 A CN 115855869A
- Authority
- CN
- China
- Prior art keywords
- tree
- tree species
- module
- species
- image
- 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
Abstract
The invention belongs to the technical field of detection of a gas phase mass spectrometer for tree species, and discloses a detection device and a detection method of the gas phase mass spectrometer for the tree species, wherein the detection device of the gas phase mass spectrometer for the tree species comprises the following steps: the device comprises a tree species chromatogram acquisition module, a main control module, a chromatographic characteristic extraction module, an analysis module, a tree species identification module, a tree species classification module, a tree species quality evaluation module and a display module. According to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved; meanwhile, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, the sample does not need to be subjected to complex pretreatment before detection, a large amount of labor force is saved, different types of tree species can be rapidly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved.
Description
Technical Field
The invention belongs to the technical field of detection of a gas-phase mass spectrometer for tree species, and particularly relates to a detection device and a detection method of the gas-phase mass spectrometer for the tree species.
Background
The gas chromatography-mass spectrometer is widely applied to the fields of environmental protection industry, electronic industry, textile industry, petrochemical industry, essence and spice industry, medicine industry, agriculture, food safety and the like; analyzing organic pollutants in the environment (analyzing air, water quality and pollution in soil); analyzing pesticide residues, animal residues and medicine residues; analyzing aroma components of the essence and the spice; detecting harmful substances in the textile industry; however, the existing detection device and method of the gas-phase mass spectrometer for the tree seeds cannot obtain deeper information inside the tree seeds, and the identification precision can be reduced; meanwhile, the tree species are not accurately classified.
Through the above analysis, the problems and defects of the prior art are as follows:
(1) The existing detection device and method of the gas phase mass spectrometer for the tree seeds cannot obtain deeper information inside the tree seeds, and the identification precision can be reduced.
(2) The accuracy of tree species classification and identification needs to be improved.
Disclosure of Invention
Aiming at the problems in the prior art, the invention provides a detection device and a detection method of a gas-phase mass spectrometer for tree species.
The invention is realized in this way, a kind of tree species uses the checkout gear of the gas phase mass spectrometer including:
the system comprises a tree species chromatogram acquisition module, a main control module, a chromatographic characteristic extraction module, an analysis module, a tree species identification module, a tree species classification module, a tree species quality evaluation module and a display module;
the tree species chromatogram acquisition module is connected with the main control module and is used for acquiring a tree species chromatogram;
the tree chromatogram acquisition module comprises a tree chromatogram shooting unit, an image preprocessing unit and an image transmission unit;
the main control module is connected with the tree chromatogram acquisition module, the chromatographic characteristic extraction module, the analysis module, the tree species identification module, the tree species classification module, the tree species quality evaluation module and the display module and is used for controlling the normal work of each module;
the chromatographic characteristic extraction module is connected with the main control module and is used for extracting the tree chromatogram characteristics;
the chromatographic feature extraction module comprises an SIFT feature extraction unit;
the analysis module is connected with the main control module and is used for analyzing the features of the tree chromatogram;
the analysis module comprises an image analysis unit;
the tree species identification module is connected with the main control module and used for training to generate a spectrum image nondestructive identification model to identify the quality of the tree species;
the tree species classification module is connected with the main control module and used for training the generation tree species classification module to classify the tree species;
the tree seed quality evaluation module is connected with the main control module and used for evaluating the tree seed quality;
the tree species quality evaluation module comprises a quality evaluation unit;
and the display module is connected with the main control module and is used for displaying the tree chromatogram, the analysis result, the identification result, the classification result and the evaluation result.
Further, the detection method of the tree species by the gas phase mass spectrometer comprises the following steps:
firstly, acquiring a tree chromatogram through a tree chromatogram camera unit in a tree chromatogram acquisition module, then preprocessing a shot image through an image preprocessing unit, wherein the preprocessing operation comprises image denoising, image enhancement and image normalization processing, unifying the shot image into a JPG format, and finally transmitting the processed image to a main control module through an image transmission unit;
secondly, the main control module transmits the tree chromatogram to a chromatographic feature extraction module, a set SIFT algorithm is utilized to extract features of the tree chromatogram by a SIFT feature extraction unit, and extracted feature information is fed back to the main control module;
thirdly, the tree species identification module and the tree species classification module respectively feed back the obtained spectrum image nondestructive identification model and the tree species classification model to the main control module; the main control module sends the characteristic information, the spectrum image nondestructive identification model and the tree species classification model interface to the analysis module, and the image analysis unit is combined with the two models to analyze the image to obtain the tree species spectrum identification result and the tree species category information;
evaluating the quality of the tree seeds through a quality evaluation unit in the tree seed quality evaluation module, and feeding back the obtained evaluation result to the main control module; and the display module receives and displays the tree species chromatogram, the analysis result, the identification result, the classification result and the evaluation result from the main control module.
Further, the tree species identification module identification method comprises the following steps:
(1) Constructing a tree seed database, and storing the existing tree seed data into the tree seed database; collecting modeling data of various types of tree seeds, wherein the modeling data of each type of tree seeds is a transmission spectrum image of the type of tree seeds when the type of tree seeds are respectively in N near-infrared light with different wavelengths;
(2) Selecting a spectral image corresponding to a wavelength from the modeling data of each category of tree seeds as a subtractive image of the category according to the chemical characteristics of the tree seeds of each category and the chemical characteristics of molecular absorption in a near infrared region, wherein the subtractive image is used for representing interference information;
(3) Respectively carrying out difference operation on the subtracted images of the category by the spectral images corresponding to other N-1 wavelengths in the modeling data of the tree seed grains of each category to obtain N-1 groups of images of the category; extracting tree seed images in the N-1 groups of images of each category to obtain N-1 groups of tree seed images of each category;
(4) Calculating the optimal characteristic data of the images of the N-1 groups of tree seeds of each category as modeling training data of the category; establishing a spectral image nondestructive identification model of the tree species by using a support vector machine method according to the modeling training data of each category; the N near infrared lights with different wavelengths are short-wave near infrared lights with the wavelengths of 700-1100 nm.
Further, the selection method of the reduced number image comprises the following steps:
and selecting the spectral image with the minimum information content representing the tree species in the modeling data of the tree species grains of each category, and taking the spectral image as the subtraction image of the category.
Further, the extracting the tree seed grain images in the N-1 sets of images for each category includes:
respectively carrying out image segmentation on the N-1 groups of images of each category by adopting a threshold segmentation algorithm, and extracting the minimum rectangular region where the tree seeds in each group of images are located;
and amplifying the minimum rectangular area where the tree seed grains are located by adopting a nearest neighbor interpolation algorithm to obtain N1 groups of tree seed grain images of each category.
Further, the calculating the optimal feature data of the images of the N-1 sets of tree seed kernels of each category as the modeling training data of the category includes:
extracting 3 image features of each tree seed image in N-1 tree seed images of each category by respectively adopting a gray level histogram image feature extraction method, a gray level distribution statistics and gray level co-occurrence matrix image feature extraction method and a local binary pattern image feature extraction method;
respectively performing feature dimensionality reduction on 3 image features of each group of tree seed images to obtain 3 feature data of each group of tree seed images in N-1 groups of tree seed images of each category;
calculating the separability of 3 feature data of each group of tree seed images in the N-1 groups of tree seed images of each category, and comparing separability indexes of the feature data in each group to obtain the optimal feature data in each group and a corresponding optimal image feature extraction method;
and comparing the separability indexes of the optimal feature data among the N-1 groups of each category to obtain the final optimal feature data of the N-1 groups of tree seed images of each category and the corresponding optimal near infrared light wavelength for acquiring the transmission spectrum data, and taking the final optimal feature data as the modeling training data of the category.
Further, the performing feature dimension reduction on 3 image features of each group of tree seed images includes:
and respectively reducing the dimensions of the 3 image characteristics of each group of tree seed images by adopting a principal component analysis method and an orthogonal linear discriminant analysis method to obtain 3 characteristic data of each group of tree seed images.
Further, the determining the optimal dimension reduced by the principal component analysis method by using a cross validation method specifically includes:
randomly distributing image features with the number of K of each group of samples into L equal parts, sequentially taking out 1 part of the L equal parts as a test set, sequentially taking the rest L-1 parts as a training set to establish a test model, sequentially increasing the dimension to be reduced preset by a principal component analysis method, and selecting the dimension when the accuracy of the test set is highest as the optimal dimension to which the principal component analysis method is reduced;
the separability index is the inter-class intra-class relative distance, and the calculation formula of the inter-class intra-class relative distance is as follows:
wherein R is ij Is the inter-class relative distance between the i-th class and the j-th class, D ij Represents the squared Euclidean distance, W, of the centroid of the ith and jth classes i Mean sum of squared deviations, W, within class i j Represents the mean sum of squared deviations within class j.
Further, the tree species classification module comprises the following classification method:
1) Obtaining a tree seed database; the tree species database comprises a plurality of pairs of tree species samples and corresponding tree species types; carrying out gas chromatography-ion mobility spectrometry detection on the tree species samples in the tree species database to obtain a characteristic sample set;
2) Performing iterative training on the multilayer perceptron according to the characteristic sample set and the corresponding tree species type to obtain a tree species classification model; carrying out gas chromatography-ion mobility spectrometry detection on the tree species to be classified to obtain the characteristics to be classified; and classifying the tree species to be classified according to the features to be classified based on the tree species classification model, and determining the category of the tree species to be classified.
Further, the obtaining the tree species database specifically includes:
weighing 0.6g of tree species sample for each tree species, placing the tree species sample in a 20mL headspace bottle, incubating the tree species sample for 26min at 36 ℃, and then sampling for 500uL to obtain a tree species database;
the gas chromatography-ion mobility spectrometry detection of the tree species samples in the tree species database to obtain a characteristic sample set specifically comprises:
performing gas chromatography-ion mobility spectrometry detection on each tree species sample by using a flavor analyzer under a set detection condition to obtain a corresponding three-dimensional spectrogram;
determining the signal peak volume corresponding to each compound in the tree species sample according to the three-dimensional spectrogram, and taking the signal peak volume as the characteristic data of the tree species sample;
determining a characteristic sample set according to the characteristic data of each tree sample;
determining a signal peak volume corresponding to each compound in the tree species sample according to the three-dimensional spectrogram, wherein the determining is used as feature data of the tree species sample, and specifically comprises the following steps:
performing qualitative and quantitative analysis on the three-dimensional spectrogram to obtain the signal peak volume corresponding to each compound in the tree species sample;
the detection conditions include: sample introduction mode, analysis time, chromatographic column type, column temperature carrier/drift gas and micro-stepping motor temperature;
the sample introduction mode is that under the set sample introduction condition, after the tree species sample is incubated, headspace sample introduction is carried out; the analysis time is 36min; the type of the chromatographic column is FS-SE-54-CB-115m ID; the column temperature is 86 ℃; the carrier gas/drift gas is nitrogen; the micro-stepper motor temperature was 46 ℃.
In combination with the technical solutions and the technical problems to be solved, please analyze the advantages and positive effects of the technical solutions to be protected in the present invention from the following aspects:
first, aiming at the technical problems existing in the prior art and the difficulty in solving the problems, the technical problems to be solved by the technical scheme of the present invention are closely combined with results, data and the like in the research and development process, and some creative technical effects are brought after the problems are solved. The specific description is as follows:
according to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved; meanwhile, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, so that different types of tree species can be quickly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved.
Secondly, considering the technical scheme as a whole or from the perspective of products, the technical effect and advantages of the technical scheme to be protected by the invention are specifically described as follows:
according to the invention, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, and the sample does not need to be subjected to complex pretreatment before detection, so that a large amount of labor force is saved, and the detection efficiency is improved. The multi-layer perceptron is trained to obtain the tree species classification model, and then the classification of the tree species to be classified is determined based on the tree species classification model, so that the accuracy of tree species type discrimination is improved.
Drawings
Fig. 1 is a flow chart of a detection method of a gas phase mass spectrometer for tree species according to an embodiment of the present invention.
Fig. 2 is a block diagram of a detection apparatus for a gas mass spectrometer for tree species according to an embodiment of the present invention.
Fig. 3 is a flowchart of an authentication method of a tree species authentication module according to an embodiment of the present invention.
Fig. 4 is a flowchart of a tree classification module classification method according to an embodiment of the present invention.
In fig. 2: 1. a tree chromatogram acquisition module; 11. a tree species chromatogram image pickup unit; 12. an image preprocessing unit; 13. an image transmission unit; 2. a main control module; 3. a chromatographic characteristic extraction module; 31. an SIFT feature extraction unit; 4. an analysis module; 41. an image analysis unit; 5. a tree species identification module; 6. a tree species classification module; 7. a tree species quality evaluation module; 71. a quality evaluation unit; 8. and a display module.
Detailed Description
In order to make the objects, technical solutions and advantages of the present invention more apparent, the present invention is further described in detail with reference to the following embodiments. It should be understood that the specific embodiments described herein are merely illustrative of the invention and are not intended to limit the invention.
1. Illustrative embodiments are explained. This section is an explanatory embodiment expanding on the claims so as to fully understand how the present invention is embodied by those skilled in the art.
As shown in fig. 1, the detection method of the gas phase mass spectrometer for tree species provided by the invention comprises the following steps:
s101, collecting a tree chromatogram through a tree chromatogram camera shooting unit in a tree chromatogram collection module, then carrying out preprocessing operations on a shot image through an image preprocessing unit, wherein the preprocessing operations comprise image denoising, image enhancement and image normalization processing, unifying the shot image into JPG (joint photographic experts group) in an image format, and finally transmitting the processed image to a main control module through an image transmission unit;
s102, the main control module transmits the tree chromatogram map to a chromatographic feature extraction module, a set SIFT algorithm is utilized to extract features of the tree chromatogram map through a SIFT feature extraction unit, and extracted feature information is fed back to the main control module;
s103, the tree species identification module and the tree species classification module respectively feed back the obtained spectrum image nondestructive identification model and the tree species classification model to the main control module; the main control module sends the characteristic information, the spectrum image nondestructive identification model and the tree species classification model interface to the analysis module, and the image analysis unit is combined with the two models to analyze the image to obtain the tree species spectrum identification result and the tree species category information;
s104, evaluating the quality of the tree seeds through a quality evaluation unit in the tree seed quality evaluation module, and feeding back the obtained evaluation result to the main control module; and the display module receives and displays the tree species chromatogram, the analysis result, the identification result, the classification result and the evaluation result from the main control module.
According to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved; meanwhile, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, the sample does not need to be subjected to complex pretreatment before detection, a large amount of labor force is saved, different types of tree species can be rapidly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved. In addition, the multi-layer perceptron is trained to obtain a tree species classification model, and then the classification of the tree species to be classified is determined based on the tree species classification model, so that the accuracy of tree species type discrimination is improved.
As shown in fig. 2, the detection apparatus for a gas mass spectrometer for tree species according to an embodiment of the present invention includes: the device comprises a tree species chromatogram acquisition module 1, a main control module 2, a chromatographic characteristic extraction module 3, an analysis module 4, a tree species identification module 5, a tree species classification module 6, a tree species quality evaluation module 7 and a display module 8.
The tree species chromatogram acquisition module 1 is connected with the main control module 2 and is used for acquiring a tree species chromatogram;
the tree species chromatogram acquisition module 1 comprises a tree species chromatogram camera unit 11, an image preprocessing unit 12 and an image transmission unit 13;
the main control module 2 is connected with the tree species chromatogram acquisition module 1, the chromatographic characteristic extraction module 3, the analysis module 4, the tree species identification module 5, the tree species classification module 6, the tree species quality evaluation module 7 and the display module 8 and is used for controlling the normal work of each module;
the chromatographic characteristic extraction module 3 is connected with the main control module 2 and is used for extracting the tree chromatogram characteristics;
the chromatographic feature extraction module 3 comprises a SIFT feature extraction unit 31;
the analysis module 4 is connected with the main control module 2 and is used for analyzing the tree chromatogram characteristics;
the analysis module 4 comprises an image analysis unit 41;
the tree species identification module 5 is connected with the main control module 2 and used for training to generate a spectrum image nondestructive identification model to identify the quality of tree species;
the tree species classification module 6 is connected with the main control module 2 and used for training the generation tree species classification module to classify the tree species;
the tree seed quality evaluation module 7 is connected with the main control module 2 and used for evaluating the quality of the tree seeds;
the tree species quality evaluation module 7 comprises a quality evaluation unit 71;
and the display module 8 is connected with the main control module 2 and is used for displaying the tree chromatogram, the analysis result, the identification result, the classification result and the evaluation result.
According to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved; meanwhile, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, the sample does not need to be subjected to complex pretreatment before detection, a large amount of labor force is saved, different types of tree species can be rapidly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved. In addition, the multi-layer perceptron is trained to obtain a tree species classification model, and then the classification of the tree species to be classified is determined based on the tree species classification model, so that the accuracy of tree species type discrimination is improved.
As shown in fig. 3, the tree species identification module 5 provided by the present invention has the following identification method:
s201, constructing a tree seed database, and storing the existing tree seed data into the tree seed database; collecting modeling data of various types of tree seeds, wherein the modeling data of each type of tree seeds is a transmission spectrum image of the type of tree seeds when the type of tree seeds are respectively in N near-infrared light with different wavelengths;
s202, selecting a spectral image corresponding to a wavelength from the modeling data of the tree seeds of each category as a subtractive image of the category according to the chemical characteristics of the tree seeds of each category and the chemical characteristics of molecular absorption in a near infrared region, wherein the subtractive image is used for representing interference information;
s203, performing difference operation on the subtracted images of the category respectively by the spectral images corresponding to other N-1 wavelengths in the modeling data of the tree seeds of each category to obtain N-1 groups of images of the category; extracting tree seed images in the N-1 groups of images of each category to obtain N-1 groups of tree seed images of each category;
s204, calculating the optimal characteristic data of the images of the N-1 groups of tree seeds of each category as modeling training data of the category; establishing a spectral image nondestructive identification model of the tree species by using a support vector machine method according to the modeling training data of each category; the N near infrared lights with different wavelengths are short-wave near infrared lights with the wavelengths of 700-1100 nm.
According to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved.
The invention provides a method for selecting a reduced number image, which comprises the following steps:
and selecting the spectral image with the minimum information content representing the tree species in the modeling data of the tree species grains of each category, and taking the spectral image as the subtraction image of the category.
The invention provides a method for extracting tree seed images in N-1 groups of images of each category, which comprises the following steps:
respectively carrying out image segmentation on the N-1 groups of images of each category by adopting a threshold segmentation algorithm, and extracting a minimum rectangular region where the tree seeds in each group of images are located;
and amplifying the minimum rectangular area where the tree seed grains are located by adopting a nearest neighbor interpolation algorithm to obtain N1 groups of tree seed grain images of each category.
The invention provides a method for calculating the optimal characteristic data of the image of the N-1 groups of tree seeds of each category as the modeling training data of the category, which comprises the following steps:
extracting 3 image features of each tree seed image in N-1 tree seed images of each category by respectively adopting a gray level histogram image feature extraction method, a gray level distribution statistics and gray level co-occurrence matrix image feature extraction method and a local binary pattern image feature extraction method;
respectively carrying out feature dimensionality reduction on 3 image features of each group of tree seed images to obtain 3 feature data of each group of tree seed images in each category of N-1 groups of tree seed images;
calculating the separability of 3 feature data of each group of tree seed images in the N-1 groups of tree seed images of each category, and comparing separability indexes of the feature data in each group to obtain the optimal feature data in each group and a corresponding optimal image feature extraction method;
and comparing the separability indexes of the optimal feature data among the N-1 groups of each category to obtain the final optimal feature data of the N-1 groups of tree seed images of each category and the corresponding optimal near infrared light wavelength for acquiring the transmission spectrum data, and taking the final optimal feature data as the modeling training data of the category.
The invention provides a method for performing feature dimension reduction on 3 image features of each group of tree seed images, which comprises the following steps:
and respectively reducing the dimensions of the 3 image characteristics of each group of tree seed images by adopting a principal component analysis method and an orthogonal linear discriminant analysis method to obtain 3 characteristic data of each group of tree seed images.
The invention provides a method for determining the optimal dimension reduced by the principal component analysis method by adopting a cross validation method, which specifically comprises the following steps:
randomly distributing image features with the number of K of each group of samples into L equal parts, sequentially taking out 1 part of the L equal parts as a test set, sequentially taking the rest L-1 parts as a training set to establish a test model, sequentially increasing the dimension to be reduced preset by a principal component analysis method, and selecting the dimension when the accuracy of the test set is highest as the optimal dimension to which the principal component analysis method is reduced;
the separability index is the inter-class intra-class relative distance, and the calculation formula of the inter-class intra-class relative distance is as follows:
wherein R is ij Is the inter-class relative distance between the i-th class and the j-th class, D ij Represents the squared Euclidean distance, W, of the centroid of the ith and jth classes i Mean squared deviation within the ith classAnd, W j Represents the mean sum of squared deviations within class j.
As shown in fig. 4, the tree species classification module 6 provided by the present invention has the following classification method:
s301, acquiring a tree seed database; the tree species database comprises a plurality of pairs of tree species samples and corresponding tree species types; carrying out gas chromatography-ion mobility spectrometry detection on the tree species samples in the tree species database to obtain a characteristic sample set;
s302, performing iterative training on the multilayer perceptron according to the characteristic sample set and the corresponding tree type to obtain a tree classification model; carrying out gas chromatography-ion mobility spectrometry detection on the tree species to be classified to obtain the characteristics to be classified; and classifying the tree species to be classified according to the features to be classified based on the tree species classification model, and determining the category of the tree species to be classified.
According to the invention, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, the sample does not need to be subjected to complex pretreatment before detection, a large amount of labor force is saved, different types of tree species can be rapidly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved. In addition, the multi-layer perceptron is trained to obtain a tree species classification model, and then the classification of the tree species to be classified is determined based on the tree species classification model, so that the accuracy of tree species type discrimination is improved.
The invention provides a tree species database acquisition method, which specifically comprises the following steps:
weighing 0.6g of tree species sample for each tree species, placing the tree species sample in a 20mL headspace bottle, incubating the tree species sample for 26min at 36 ℃, and then sampling for 500uL to obtain a tree species database;
the gas chromatography-ion mobility spectrometry detection of the tree species samples in the tree species database to obtain a characteristic sample set specifically comprises:
performing gas chromatography-ion mobility spectrometry detection on each tree species sample by using a flavor analyzer under a set detection condition to obtain a corresponding three-dimensional spectrogram;
determining the signal peak volume corresponding to each compound in the tree species sample according to the three-dimensional spectrogram, and taking the signal peak volume as the characteristic data of the tree species sample;
determining a characteristic sample set according to the characteristic data of each tree sample;
determining a signal peak volume corresponding to each compound in the tree species sample according to the three-dimensional spectrogram, wherein the determining is used as feature data of the tree species sample, and specifically comprises the following steps:
performing qualitative and quantitative analysis on the three-dimensional spectrogram to obtain the signal peak volume corresponding to each compound in the tree species sample;
the detection conditions include: sample introduction mode, analysis time, chromatographic column type, column temperature carrier/drift gas and micro-stepping motor temperature;
the sample introduction mode is that under the set sample introduction condition, the headspace sample introduction is carried out after the tree species sample is incubated; the analysis time is 36min; the type of the chromatographic column is FS-SE-54-CB-115m ID; the column temperature is 86 ℃; the carrier/drift gas is nitrogen; the micro-stepper motor temperature was 46 ℃.
2. Application examples. In order to prove the creativity and the technical value of the technical scheme of the invention, the part is the application example of the technical scheme of the claims on specific products or related technologies.
According to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved; meanwhile, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, the sample does not need to be subjected to complex pretreatment before detection, a large amount of labor force is saved, different types of tree species can be quickly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved. In addition, the multi-layer perceptron is trained to obtain a tree species classification model, and then the category of the tree species to be classified is determined based on the tree species classification model, so that the accuracy of tree species type discrimination is improved.
It should be noted that embodiments of the present invention can be realized in hardware, software, or a combination of software and hardware. The hardware portion may be implemented using dedicated logic; the software portions may be stored in a memory and executed by a suitable instruction execution system, such as a microprocessor or specially designed hardware. Those skilled in the art will appreciate that the apparatus and methods described above may be implemented using computer executable instructions and/or embodied in processor control code, such code being provided on a carrier medium such as a disk, CD-or DVD-ROM, programmable memory such as read only memory (firmware), or a data carrier such as an optical or electronic signal carrier, for example. The apparatus and its modules of the present invention may be implemented by hardware circuits such as very large scale integrated circuits or gate arrays, semiconductors such as logic chips, transistors, or programmable hardware devices such as field programmable gate arrays, programmable logic devices, etc., or by software executed by various types of processors, or by a combination of hardware circuits and software, e.g., firmware.
3. Evidence of the relevant effects of the examples. The embodiment of the invention achieves some positive effects in the process of research and development or use, and has great advantages compared with the prior art, and the following contents are described by combining data, diagrams and the like in the test process.
According to the invention, the tree seed identification module adopts a transmission imaging mode, so that complete internal information of tree seeds can be acquired, the influence of the position effect of the tree seeds is overcome, the properties of the tree seeds can be accurately analyzed, and the identification precision is improved; meanwhile, the tree species classification module is used for detecting the sample to be detected by adopting the gas chromatography-ion mobility spectrometry, the sample does not need to be subjected to complex pretreatment before detection, a large amount of labor force is saved, different types of tree species can be rapidly detected, the gas chromatography-ion mobility spectrometry detection method is high in resolution and sensitivity, and the detection efficiency is improved. In addition, the multi-layer perceptron is trained to obtain a tree species classification model, and then the classification of the tree species to be classified is determined based on the tree species classification model, so that the accuracy of tree species type discrimination is improved.
The above description is only for the purpose of illustrating the embodiments of the present invention, and the scope of the present invention should not be limited thereto, and any modifications, equivalents and improvements made by those skilled in the art within the technical scope of the present invention as disclosed in the present invention should be covered by the scope of the present invention.
Claims (10)
1. A detection device for a gas phase mass spectrometer for tree species is characterized by comprising:
the system comprises a tree species chromatogram acquisition module, a main control module, a chromatographic characteristic extraction module, an analysis module, a tree species identification module, a tree species classification module, a tree species quality evaluation module and a display module;
the tree species chromatogram acquisition module is connected with the main control module and is used for acquiring a tree species chromatogram;
the tree chromatogram acquisition module comprises a tree chromatogram camera unit, an image preprocessing unit and an image transmission unit;
the main control module is connected with the tree chromatogram acquisition module, the chromatographic characteristic extraction module, the analysis module, the tree species identification module, the tree species classification module, the tree species quality evaluation module and the display module and is used for controlling the normal work of each module;
the chromatographic characteristic extraction module is connected with the main control module and is used for extracting the tree chromatogram characteristics;
the chromatographic feature extraction module comprises an SIFT feature extraction unit;
the analysis module is connected with the main control module and is used for analyzing the characteristics of the tree chromatogram;
the analysis module comprises an image analysis unit;
the tree species identification module is connected with the main control module and used for training to generate a spectrum image nondestructive identification model to identify the quality of the tree species;
the tree species classification module is connected with the main control module and used for training the generation tree species classification module to classify the tree species;
the tree seed quality evaluation module is connected with the main control module and used for evaluating the tree seed quality;
the tree species quality evaluation module comprises a quality evaluation unit;
and the display module is connected with the main control module and is used for displaying the tree chromatogram, the analysis result, the identification result, the classification result and the evaluation result.
2. The method for detecting a gas phase mass spectrometer for tree species according to claim 1, wherein the method for detecting a gas phase mass spectrometer for tree species comprises the following steps:
firstly, acquiring a tree chromatogram through a tree chromatogram camera unit in a tree chromatogram acquisition module, then preprocessing a shot image through an image preprocessing unit, wherein the preprocessing operation comprises image denoising, image enhancement and image normalization processing, unifying the shot image into a JPG format, and finally transmitting the processed image to a main control module through an image transmission unit;
secondly, the main control module transmits the tree chromatogram to a chromatographic feature extraction module, a set SIFT algorithm is utilized to extract features of the tree chromatogram by a SIFT feature extraction unit, and extracted feature information is fed back to the main control module;
thirdly, the tree species identification module and the tree species classification module respectively feed back the obtained spectrum image nondestructive identification model and the tree species classification model to the main control module; the main control module sends the characteristic information, the spectrum image nondestructive identification model and the tree species classification model interface to the analysis module, and the image analysis unit is combined with the two models to analyze the image to obtain the tree species spectrum identification result and the tree species category information;
evaluating the quality of the tree seeds through a quality evaluation unit in the tree seed quality evaluation module, and feeding back the obtained evaluation result to the main control module; and the display module receives and displays the tree chromatogram, the analysis result, the identification result, the classification result and the evaluation result from the main control module.
3. The apparatus for detecting species by gas mass spectrometer as claimed in claim 1, wherein said species identification module is configured to identify species by:
(1) Constructing a tree seed database, and storing the existing tree seed data into the tree seed database; collecting modeling data of various types of tree seeds, wherein the modeling data of each type of tree seeds is a transmission spectrum image of the type of tree seeds when the type of tree seeds are respectively in N near-infrared light with different wavelengths;
(2) Selecting a spectral image corresponding to a wavelength from the modeling data of each category of tree seeds as a subtractive image of the category according to the chemical characteristics of the tree seeds of each category and the chemical characteristics of molecular absorption in a near infrared region, wherein the subtractive image is used for representing interference information;
(3) Respectively carrying out difference operation on the subtracted images of the category by the spectral images corresponding to other N-1 wavelengths in the modeling data of the tree seed grains of each category to obtain N-1 groups of images of the category; extracting tree seed images in the N-1 groups of images of each category to obtain N-1 groups of tree seed images of each category;
(4) Calculating the optimal characteristic data of the images of the N-1 groups of tree seeds of each category as modeling training data of the category; establishing a spectral image nondestructive identification model of the tree species by using a support vector machine method according to the modeling training data of each category; the N near infrared lights with different wavelengths are short-wave near infrared lights with the wavelengths of 700-1100 nm.
4. The apparatus for detecting a gas phase mass spectrometer for species of claim 3, wherein said subtraction image is selected by:
and selecting the spectral image with the minimum information content representing the tree species in the modeling data of the tree species grains of each category, and taking the spectral image as the subtraction image of the category.
5. The apparatus for detecting a gas phase mass spectrometer for tree seeds of claim 3, wherein said extracting the tree seed grain image in the N-1 sets of images for each category comprises:
respectively carrying out image segmentation on the N-1 groups of images of each category by adopting a threshold segmentation algorithm, and extracting the minimum rectangular region where the tree seeds in each group of images are located;
and amplifying the minimum rectangular area where the tree seed grains are located by adopting a nearest neighbor interpolation algorithm to obtain N1 groups of tree seed grain images of each category.
6. The apparatus for detecting a gas phase mass spectrometer for tree seeds as claimed in claim 3, wherein said calculating the optimal feature data of the image of the N-1 sets of tree seed kernels of each category as the modeling training data of the category comprises:
extracting 3 image features of each tree seed image in N-1 tree seed images of each category by respectively adopting a gray level histogram image feature extraction method, a gray level distribution statistics and gray level co-occurrence matrix image feature extraction method and a local binary pattern image feature extraction method;
respectively performing feature dimensionality reduction on 3 image features of each group of tree seed images to obtain 3 feature data of each group of tree seed images in N-1 groups of tree seed images of each category;
calculating the separability of 3 feature data of each group of tree seed images in N-1 groups of tree seed images of each category, and comparing separability indexes of the feature data in each group to obtain optimal feature data in each group and a corresponding optimal image feature extraction method;
and comparing the separability indexes of the optimal feature data among the N-1 groups of each category to obtain the final optimal feature data of the N-1 groups of tree seed images of each category and the corresponding optimal near infrared light wavelength for acquiring the transmission spectrum data, and taking the final optimal feature data as the modeling training data of the category.
7. The apparatus for detecting a tree species using a gas phase mass spectrometer as claimed in claim 6, wherein said performing feature dimensionality reduction on 3 image features of each set of tree seed images comprises:
and respectively reducing the dimensions of the 3 image characteristics of each group of tree seed images by adopting a principal component analysis method and an orthogonal linear discriminant analysis method to obtain 3 characteristic data of each group of tree seed images.
8. The apparatus for detecting a gas phase mass spectrometer for tree species of claim 3, wherein said determining the optimal dimension to which said principal component analysis is reduced by using a cross-validation method comprises:
randomly distributing image features with the number of K of each group of samples into L equal parts, sequentially taking out 1 part of the L equal parts as a test set, sequentially taking the rest L-1 parts as a training set to establish a test model, sequentially increasing the dimension to be reduced preset by a principal component analysis method, and selecting the dimension when the accuracy of the test set is highest as the optimal dimension to which the principal component analysis method is reduced;
the separability index is the inter-class intra-class relative distance, and the calculation formula of the inter-class intra-class relative distance is as follows:
wherein R is ij Is the inter-class relative distance between the i-th class and the j-th class, D ij Represents the squared Euclidean distance, W, of the center of gravity of the ith and jth classes i Mean sum of squared deviations, W, within class i j Represents the mean sum of squared deviations within class j.
9. The apparatus for detecting species by gas mass spectrometer as claimed in claim 1, wherein said species classification module classifies the species by the following method:
1) Obtaining a tree seed database; the tree species database comprises a plurality of pairs of tree species samples and corresponding tree species types; carrying out gas chromatography-ion mobility spectrometry detection on the tree species samples in the tree species database to obtain a characteristic sample set;
2) Performing iterative training on the multilayer perceptron according to the characteristic sample set and the corresponding tree species type to obtain a tree species classification model; carrying out gas chromatography-ion mobility spectrometry detection on the tree species to be classified to obtain the characteristics to be classified; and classifying the tree species to be classified according to the features to be classified based on the tree species classification model, and determining the category of the tree species to be classified.
10. The apparatus for detecting a gas mass spectrometer for species as claimed in claim 9, wherein said obtaining the database of species comprises:
weighing 0.6g of tree species sample for each tree species, placing the tree species sample in a 20mL headspace bottle, incubating the tree species sample for 26min at 36 ℃, and then sampling for 500uL to obtain a tree species database;
the gas chromatography-ion mobility spectrometry detection of the tree species samples in the tree species database to obtain a characteristic sample set specifically comprises:
for each tree species sample, carrying out gas chromatography-ion mobility spectrometry detection on the tree species sample by using a flavor analyzer under a set detection condition to obtain a corresponding three-dimensional spectrogram;
determining the signal peak volume corresponding to each compound in the tree species sample according to the three-dimensional spectrogram, and taking the signal peak volume as the characteristic data of the tree species sample;
determining a characteristic sample set according to the characteristic data of each tree sample;
determining a signal peak volume corresponding to each compound in the tree species sample according to the three-dimensional spectrogram, wherein the determining is used as feature data of the tree species sample, and specifically comprises the following steps:
performing qualitative and quantitative analysis on the three-dimensional spectrogram to obtain the signal peak volume corresponding to each compound in the tree species sample;
the detection conditions include: sample introduction mode, analysis time, chromatographic column type, column temperature carrier/drift gas and micro-stepping motor temperature;
the sample introduction mode is that under the set sample introduction condition, the headspace sample introduction is carried out after the tree species sample is incubated; the analysis time is 36min; the type of the chromatographic column is FS-SE-54-CB-115m ID; the column temperature is 86 ℃; the carrier gas/drift gas is nitrogen; the micro-stepper motor temperature was 46 ℃.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211447444.0A CN115855869A (en) | 2022-11-18 | 2022-11-18 | Detection device and method of gas-phase mass spectrometer for tree species |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202211447444.0A CN115855869A (en) | 2022-11-18 | 2022-11-18 | Detection device and method of gas-phase mass spectrometer for tree species |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115855869A true CN115855869A (en) | 2023-03-28 |
Family
ID=85664169
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202211447444.0A Pending CN115855869A (en) | 2022-11-18 | 2022-11-18 | Detection device and method of gas-phase mass spectrometer for tree species |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN115855869A (en) |
Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102914597A (en) * | 2011-08-02 | 2013-02-06 | 漳州片仔癀药业股份有限公司 | Quality testing method for fingerprint of herbal medicine musk |
| CN104297504A (en) * | 2014-10-22 | 2015-01-21 | 上海申腾信息技术有限公司 | Automatic gas chromatographic control system |
| CN104990892A (en) * | 2015-06-24 | 2015-10-21 | 中国农业大学 | Spectrum image lossless identification model establishing method for seeds and seed identification method |
| CN109061020A (en) * | 2018-09-28 | 2018-12-21 | 深圳市绘云生物科技有限公司 | A kind of data analysis system based on gas phase and liquid phase chromatographic mass spectrometry platform |
| CN109781917A (en) * | 2017-11-14 | 2019-05-21 | 中国科学院大连化学物理研究所 | A kind of biological sample intelligent identification Method based on molecule map |
| CN110244009A (en) * | 2019-06-18 | 2019-09-17 | 六盘水市食品药品检验检测所 | A kind of quick identification detection system and method for honey quality |
| CN111898649A (en) * | 2020-07-07 | 2020-11-06 | 西南林业大学 | Wood tree species classification and identification method based on visible light or near infrared spectrum analysis |
| CN113899826A (en) * | 2021-09-29 | 2022-01-07 | 中国农业大学 | Method and system for classifying astragalus seeds |
-
2022
- 2022-11-18 CN CN202211447444.0A patent/CN115855869A/en active Pending
Patent Citations (8)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102914597A (en) * | 2011-08-02 | 2013-02-06 | 漳州片仔癀药业股份有限公司 | Quality testing method for fingerprint of herbal medicine musk |
| CN104297504A (en) * | 2014-10-22 | 2015-01-21 | 上海申腾信息技术有限公司 | Automatic gas chromatographic control system |
| CN104990892A (en) * | 2015-06-24 | 2015-10-21 | 中国农业大学 | Spectrum image lossless identification model establishing method for seeds and seed identification method |
| CN109781917A (en) * | 2017-11-14 | 2019-05-21 | 中国科学院大连化学物理研究所 | A kind of biological sample intelligent identification Method based on molecule map |
| CN109061020A (en) * | 2018-09-28 | 2018-12-21 | 深圳市绘云生物科技有限公司 | A kind of data analysis system based on gas phase and liquid phase chromatographic mass spectrometry platform |
| CN110244009A (en) * | 2019-06-18 | 2019-09-17 | 六盘水市食品药品检验检测所 | A kind of quick identification detection system and method for honey quality |
| CN111898649A (en) * | 2020-07-07 | 2020-11-06 | 西南林业大学 | Wood tree species classification and identification method based on visible light or near infrared spectrum analysis |
| CN113899826A (en) * | 2021-09-29 | 2022-01-07 | 中国农业大学 | Method and system for classifying astragalus seeds |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110110743B (en) | Automatic recognition system and method for seven-class mass spectrum | |
| Feilhauer et al. | Multi-method ensemble selection of spectral bands related to leaf biochemistry | |
| Li et al. | Apple quality identification and classification by image processing based on convolutional neural networks | |
| WO2018121122A1 (en) | Raman spectroscopy detection method for checking goods, and electronic device | |
| Bertani et al. | Optical detection of aflatoxins B in grained almonds using fluorescence spectroscopy and machine learning algorithms | |
| Wu et al. | Identification and quantification of counterfeit sesame oil by 3D fluorescence spectroscopy and convolutional neural network | |
| Nawaz et al. | A robust deep learning approach for tomato plant leaf disease localization and classification | |
| Lin et al. | Rapidly and exactly determining postharvest dry soybean seed quality based on machine vision technology | |
| Kumar et al. | Deep remote sensing methods for methane detection in overhead hyperspectral imagery | |
| Yin et al. | Non-destructive detection of foreign contaminants in toast bread with near infrared spectroscopy and computer vision techniques | |
| CN113899702A (en) | Quantum Fourier transform-based vaccine multispectral rapid detection method | |
| Hu et al. | Detecting different pesticide residues on Hami melon surface using hyperspectral imaging combined with 1D-CNN and information fusion | |
| Ma et al. | Neural network in food analytics | |
| Lotter et al. | Identifying plastics with photoluminescence spectroscopy and machine learning | |
| Noshiri et al. | A comprehensive review of 3D convolutional neural network-based classification techniques of diseased and defective crops using non-UAV-based hyperspectral images | |
| Tu et al. | AIseed: An automated image analysis software for high-throughput phenotyping and quality non-destructive testing of individual plant seeds | |
| Fursov et al. | Thematic classification with support subspaces in hyperspectral images | |
| CN115855869A (en) | Detection device and method of gas-phase mass spectrometer for tree species | |
| Ruiz‐Munoz et al. | Super resolution for root imaging | |
| Huang et al. | Robust and Accurate Classification of Mutton Adulteration Under Food Additives Effect Based on Multi-Part Depth Fusion Features and Optimized Support Vector Machine | |
| Li et al. | Application of laser-induced breakdown spectroscopy coupled with spectral matrix and convolutional neural network for identifying geographical origins of Gentiana rigescens Franch | |
| CN111595802A (en) | Construction method and application of Clinacanthus nutans seed source place classification model based on NIR (near infrared spectroscopy) | |
| CN111222543A (en) | Substance identification method and apparatus, and computer-readable storage medium | |
| CN107664623B (en) | Method for extracting spectral characteristics of substance | |
| Cao | Investigation of a convolutional neural network-based approach for license plate detection |
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 |