A method and apparatus for correcting alcohol concentration information

Abstract

The application provides an alcohol concentration information correction method and device. The alcohol concentration information correction method comprises the following steps: obtaining alcohol concentration information detected by an electronic nose at a current moment; obtaining a correction model; and generating corrected alcohol concentration information according to the alcohol concentration information detected by the electronic nose at the current moment and the correction model. The alcohol concentration information correction method provided by the application uses a correction model to correct alcohol concentration information detected by an electronic nose, thereby avoiding the problem of detection data errors caused by temperature changes. The problem that the low-temperature limitation cannot play a role when the electronic nose is applied to a car for alcohol driving detection is solved, and the real-time monitoring work of alcohol driving of online car hailing and shared cars by using the electronic nose under extremely cold conditions is effectively assisted, so that higher economic benefits can be obtained for the car service industry.

Description

TECHNICAL FIELD

[0001] The one or more embodiments of the specification relate to the technical field of alcohol detection, and in particular to an alcohol concentration information correction method and device. BACKGROUND

[0002] In recent years, the emergence of shared cars has promoted people's daily travel. However, avoiding drunk driving of shared cars by drivers has become a problem to be solved in the development of car sharing industry. Usually, traffic law enforcement personnel conducts handheld alcohol detection on drivers by observing driving state. However, the traditional manual detection to monitor the drinking state of drivers needs a large amount of human resources, causing waste of resources, low detection efficiency, and insufficient range. Moreover, if the state of the driver is not monitored, in this case, once the drunk driver drives the online car or shared car on the road, it will constitute a potential accident, which will have a destructive impact on the society, online car companies and shared car companies. The electronic nose has specific sensors and identification modules, which can quickly provide the overall information of the detected sample, and is very suitable for detecting odors in confined spaces such as space stations, space shuttles and cars. At present, there have been reports that the electronic nose is applied to the car to detect drunk driving.

[0003] Since the core device of the electronic nose is a gas sensor, and the most widely used is a metal oxide semiconductor sensor. The drift caused by the change of the surrounding environment and the change of the temperature of the sensor has always been difficult to analyze and eliminate, and if the sensor signal is not corrected for drift, it will affect the accuracy of the model. When the sensor is applied to the electronic nose, the interference of temperature on the gas sensor is more significant, and the temperature of the sample and the environment will cause classification errors of the electronic nose. In many countries around the world, the temperature will reach several tens of degrees below zero in winter, and the temperature in northeast China will also reach below-20℃ in winter. In this case, it is very important to process the detection data of the electronic nose.

[0004] Therefore, it is desirable to have a technical solution to overcome or at least alleviate at least one of the aforementioned deficiencies of the prior art. SUMMARY

[0005] Therefore, it is desirable to have a technical solution to overcome or at least alleviate at least one of the aforementioned deficiencies of the prior art.

[0006] In order to achieve the above purpose, one or more embodiments of the specification provide an alcohol concentration information correction method, which comprises:

[0007] obtaining alcohol concentration information detected by an electronic nose at a current time;

[0008] obtaining a correction model;

[0009] According to the alcohol concentration information detected by the electronic nose at the current time and the correction model, corrected alcohol concentration information is generated.

[0010] Optionally, before the correction model is generated, the method further comprises:

[0011] Generating a correction model.

[0012] Optionally, the generating of the correction model comprises:

[0013] Obtaining a plurality of alcohol concentration sample information collected by the electronic nose in each preset environment;

[0014] Processing all the alcohol concentration sample information using an ICA model to obtain the correction model.

[0015] Optionally, the obtaining of the plurality of alcohol concentration sample information collected by the electronic nose in each preset environment comprises:

[0016] Obtaining a plurality of alcohol concentration sample information collected by the electronic nose in a space with a preset temperature and a preset alcohol concentration.

[0017] Optionally, the processing of all the alcohol concentration sample information using an ICA model to obtain the correction model comprises:

[0018] Obtaining feature information of each alcohol concentration sample information;

[0019] Processing the feature information of all the alcohol concentration sample information using an ICA model to obtain ICA model independent components;

[0020] Training the obtained ICA model independent components using a multi-mode recognition algorithm to obtain the correction model.

[0021] Optionally, the obtaining of the feature information of each alcohol concentration sample information comprises:

[0022] Standardizing each alcohol concentration sample information;

[0023] De-noising each standardized alcohol concentration sample information;

[0024] Obtaining feature information corresponding to each de-noised alcohol concentration sample information according to the alcohol concentration sample information.

[0025] Optionally, the de-noising of each standardized alcohol concentration sample information comprises:

[0026] Using a 10Hz FFT filter to de-noise each standardized alcohol concentration sample information.

[0027] Optionally, the feature information of all the alcohol concentration sample information is analyzed by using the ICA model to obtain ICA model independent components, including:

[0028] A feature matrix is constructed according to the feature information of all the alcohol concentration sample information.

[0029] The feature matrix is subjected to mean value elimination processing.

[0030] The feature matrix subjected to the mean value elimination processing is subjected to whitening.

[0031] The ICA model independent components are obtained according to the feature matrix subjected to the mean value elimination processing and the whitening result.

[0032] Optionally, the ICA model independent components obtained are trained by using a recognition method to obtain a correction model, including:

[0033] The ICA model independent components are trained by using a support vector machine to obtain the correction model.

[0034] In another aspect, the present application also provides an alcohol concentration information correction device, including:

[0035] An alcohol concentration information acquisition module is configured to acquire alcohol concentration information detected by an electronic nose at a current time.

[0036] A correction model acquisition module is configured to acquire a correction model.

[0037] An alcohol concentration information correction module is configured to generate corrected alcohol concentration information according to the alcohol concentration information detected by the electronic nose at the current time and the correction model.

[0038] The present application has the following beneficial effects:

[0039] The alcohol concentration information correction method provided by the present application uses a correction model to correct alcohol concentration information detected by an electronic nose, thereby avoiding the problem of detection data errors caused by temperature changes. The problem that the low-temperature limitation cannot play a role when the electronic nose is applied to a car for alcohol driving detection is solved, which effectively assists the real-time monitoring of alcohol driving of online car hailing and shared cars by using the electronic nose under extremely cold conditions, and can obtain higher economic benefits for the car service industry. BRIEF DESCRIPTION OF DRAWINGS

[0040] In order to more clearly illustrate the technical solutions in the one or more embodiments of the present specification or the prior art, the drawings needed to be used in the embodiment or prior art description will be briefly introduced as follows. Obviously, the drawings in the following description only constitute one or more embodiments of the present specification, and other drawings can also be obtained by those skilled in the art without creative effort on the basis of these drawings.

[0041] Figure 1 A flowchart of a method for correcting alcohol concentration information according to one or more embodiments of the present specification is shown in the figure.

[0042] Figure 2 A schematic diagram of a data acquisition system according to one or more embodiments of the present specification is shown in the figure.

[0043] Figure 3 A more specific schematic diagram of an electronic device hardware structure according to one or more embodiments of the present specification is shown in the figure.

[0044] Figure 4 A flowchart of a method for establishing a contrast model according to one or more embodiments of the present specification is shown in the figure. DETAILED DESCRIPTION

[0045] In order to make the purposes, technical solutions and advantages of the present disclosure clearer, the present disclosure will be further described in detail below with reference to specific embodiments and drawings.

[0046] It should be noted that, unless otherwise defined, the technical terms or scientific terms used in one or more embodiments of the present specification should be understood as the usual meaning understood by those skilled in the art to which the present disclosure belongs. The terms "first", "second" and the like used in one or more embodiments of the present specification do not represent any order, number or importance, but are only used to distinguish different components. The terms "include" or "contain" and the like mean that the elements or objects before the terms cover the elements or objects listed after the terms and their equivalents, and do not exclude other elements or objects. The terms "connect" or "connected" and the like are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. The terms "up", "down", "left", "right" and the like are only used to represent relative positional relationships, and when the absolute positions of the described objects change, the relative positional relationships may also change accordingly.

[0047] Figure 1 A flowchart of a method for correcting alcohol concentration information according to one or more embodiments of the present specification is shown in the figure.

[0048] Reference Figure 1The application provides an alcohol concentration information correction method, which comprises the following steps:

[0049] obtaining alcohol concentration information detected by an electronic nose at the current moment;

[0050] obtaining a correction model;

[0051] generating corrected alcohol concentration information according to the alcohol concentration information detected by the electronic nose at the current moment and the correction model.

[0052] The alcohol concentration information correction method provided by the application uses a correction model to correct alcohol concentration information detected by an electronic nose, thereby avoiding the problem of detection data errors caused by temperature changes. The problem that the low-temperature limitation cannot play a role when the electronic nose is applied to a car for drunk driving detection is solved, and the real-time monitoring of drunk driving of online car hailing and shared cars by using the electronic nose under extremely cold conditions is effectively assisted, thereby enabling the car service industry to obtain higher economic benefits.

[0053] In an embodiment, before the correction model is obtained, the alcohol concentration information correction method further comprises the following steps:

[0054] generating the correction model.

[0055] In an embodiment, the generation of the correction model comprises the following steps:

[0056] obtaining a plurality of alcohol concentration sample information collected by the electronic nose in each preset environment;

[0057] processing all the alcohol concentration sample information by using an ICA model to obtain the correction model.

[0058] In an embodiment, the obtaining of the plurality of alcohol concentration sample information collected by the electronic nose in each preset environment comprises the following steps:

[0059] obtaining a plurality of alcohol concentration sample information collected by the electronic nose in a space with a preset temperature and a preset alcohol concentration.

[0060] In an embodiment, the processing of all the alcohol concentration sample information by using the ICA model to obtain the correction model comprises the following steps:

[0061] obtaining feature information of each alcohol concentration sample information;

[0062] analyzing the feature information of all the alcohol concentration sample information by using the ICA model to obtain ICA model independent components;

[0063] training the obtained ICA model independent components by using a recognition method to obtain the correction model.

[0064] Specifically, the recognition method can be one or more of various pattern recognition algorithms, such as random forest, support vector machine, etc.

[0065] In an embodiment, the feature information of each alcohol concentration sample information comprises:

[0066] Each alcohol concentration sample information is standardized;

[0067] Each standardized alcohol concentration sample information is denoised;

[0068] The feature information corresponding to each denoised alcohol concentration sample information is obtained.

[0069] In an embodiment, the denoising of each standardized alcohol concentration sample information comprises:

[0070] Each standardized alcohol concentration sample information is denoised using a 10Hz FFT filter.

[0071] In an embodiment, the feature information of all alcohol concentration sample information is analyzed using an ICA model to obtain ICA model independent components, comprising:

[0072] A feature matrix is constructed from the feature information of all alcohol concentration sample information;

[0073] The feature matrix is de-meaned;

[0074] The de-meaned feature matrix is whitened;

[0075] The ICA model independent components are obtained according to the de-meaned feature matrix and the whitening result.

[0076] In an embodiment, the ICA model independent components obtained are trained using a recognition method to obtain a correction model, comprising:

[0077] The ICA model independent components are trained using a support vector machine to obtain a correction model.

[0078] The alcohol concentration information correction method provided by the present application can accurately identify different concentration alcohol exhaled air in an ultra-low temperature environment without the need for a large number of training models, eliminating the negative effects of low temperature on metal oxide semiconductors and significantly improving detection accuracy.

[0079] The step of generating a correction model in the alcohol concentration information correction method provided by the present application will be further described in the following example manner. It can be understood that this example does not constitute any limitation on the present application.

[0080] First, the data collection environment information is described:

[0081] Referring to Figure 2 , the electronic nose for collecting data includes a bionic chamber, a gas flow meter, a temperature and humidity sensor, a conditioning circuit board (for filtering noise and providing power to the sensor array), a data acquisition card, a gas pump, and a metal oxide semiconductor sensor array; the sampling time is 60s, the sampling frequency is 100HZ, and the gas flow in the bionic chamber is 1L / min; the bionic chamber contains an array of 32 metal oxide semiconductor sensors, with specific models being: TGS2611, TGS2620, TGS2603, TGS2602, TGS2610, TGS2600, GSBT11, MS1100, MP135, MP901, MP-9, MP-3B, MP-4, MP-5, MP-2, MP503, MP801, MP905, MP402, WSP1110, WSP2110, WSP7110, MP-7, MP702, TGS2618-COO;

[0082] Obtaining multiple alcohol concentration sample information collected by the electronic nose in each preset environment includes the following steps:

[0083] The alcohol concentration in the exhaled gas of the simulated driver driving under the influence of alcohol is 0.1mg / L, and contains 5% carbon dioxide, 16% oxygen, and the rest is nitrogen, denoted as C1; the alcohol concentration in the exhaled gas of the simulated driver driving under the influence of alcohol is 0.5mg / L, and contains 5% carbon dioxide, 16% oxygen, and the rest is nitrogen, denoted as C2;

[0084] The temperature gradient is divided into T1=20±2℃, T2=-10±2℃, and T3=-20±2℃;

[0085] The electronic nose is placed in a laboratory incubator for environmental temperature regulation, so that their temperature is reduced to the target temperature. When the temperature reaches the target temperature, two C1 and C2 mixed gases with different alcohol concentrations are released, and the electronic nose is used to collect gas information. Each concentration sample is measured 40 times at the same temperature;

[0086] The odor data detected by the electronic nose under T1, T2, and T3 temperature conditions for C1 and C2 are divided into 6 odor sample sets according to temperature and concentration, denoted as {L C1T1 , L C1T2 , L C1T3 , L C2T1 , L C2T2 , L C2T3};

[0087] Standardizing each alcohol concentration sample information, including the following steps:

[0088] Standardizing the original data obtained by the 32 gas sensor arrays to 0-1;

[0089] The standardization formula is: Where x i is the i-th element of the original vector (input or target), x min is the minimum value in the original vector, x max is the maximum value in the original vector, y i is the i-th element of the generated vector data;

[0090] De-noising each alcohol concentration sample information after standardization, including the following steps:

[0091] Using a 10Hz FFT filter to de-noise the data after standardization. The de-noised data is each data after pre-processing, denoted as d i , and the data samples are denoted as {G 11 , G 12 , G 12 , G 21 , G 22 , G 23};

[0092] According to each alcohol concentration sample information after de-noising, the feature information corresponding to the alcohol concentration sample information is obtained, and the extraction formula is:

[0093] A feature matrix is constructed for the feature information of all alcohol concentration sample information, and the feature matrix formed by all odor samples is denoted as F, with a feature dimension of 32 (since 32 sensors are used for data collection, the feature dimension here is 32);

[0094] De-meaning the feature matrix, including the following steps:

[0095] The de-meaning formula is: f i ′ = f i - μ; where μ is the mean of f i , and the de-meaned feature matrix is denoted as F';

[0096] Whitening the feature matrix after de-meaning, including the following steps:

[0097] The whitening formula is: Z = BF'; where B is the whitening matrix;

[0098] According to the feature matrix after de-meaning and the whitening result, an ICA model independent component is obtained, including the following steps:

[0099] According to F' = AS, S = A -1 F' = WF', solve the demixing matrix W; wherein, A is an unknown mixing matrix, S is an unknown m (m≤N) dimensional signal source;

[0100] Get independent components U = WZ; wherein, U is the component removed from the strong correlation with the environmental temperature interference by using the independent component analysis (ICA) model; the embodiment removes the ICA output component strongly correlated with the environmental temperature interference to obtain the ICA component more correlated with the real odor signal.

[0101] The selected ICA component U is trained and classified by using a support vector machine (SVM) pattern recognition algorithm to establish a correction model (ICA temperature correction model).

[0102] So far, the generation of the correction model provided in the example is introduced, and in order to serve as a comparison, the following introduces the use of a support vector machine (SVM) as a classifier, and a single temperature model and two mixed temperature models are also established for comparison with the ICA temperature correction model, and the model establishment process is as shown in Figure 4 ;

[0103] 1. Single temperature model establishment: select the data at T1 temperature as the training set, denoted as K1; use the support vector machine classifier to test the prediction results at different temperatures, that is, use the data at T2 and T3 nearby temperature as the test set to classify the alcohol concentration in the gas.

[0104] 2. Establishment of two mixed temperature models, the steps are as follows:

[0105] 2.1 Use the alcohol mixed gas data of C1 and C2 at T1 and T3 temperatures as the mixed temperature data set, use the support vector machine as the classifier to establish the first mixed temperature model, and test the classification ability of the model for different alcohol concentrations of mixed gas at different temperatures. Among them, the data of T1 and T3 are used as the training set, denoted as K2; the data of T1, T2 and T3 are used as the test set, and the proportion of T1 and T3 in the training set and the test set is 7:3;

[0106] 2.2 Use the alcohol mixed gas data of C1 and C2 at T1, T2 and T3 temperatures as the mixed temperature data set, use the support vector machine as the classifier to establish the second mixed temperature model, and test the classification ability of the model for different alcohol concentrations of mixed gas at different temperatures. Among them, the data of T1, T2 and T3 are used as the training set, denoted as K3; the data of T1, T2 and T3 are used as the test set, and the proportion of T1, T2 and T3 in the training set and the test set is 7:3;

[0107] 3. The classification results of the four models are shown in the following table:

[0108]

[0109]

[0110] It can be seen from the above table that the same 240 (3 temperatures, 2 concentrations, 40 groups of data under each environment, that is, 3*2*40) groups of data are used, and the results of other training models are much lower than the method provided in the application. At the same time, it can be known that if the same results as the ICA correction model are to be obtained, the number of samples required by the single-temperature model and the two mixed-temperature models will be very large, and it can be concluded that the ICA temperature correction model can reduce the sample requirement while effectively reducing the influence of low temperature on the electronic nose to detect different concentrations of alcohol gas.

[0111] On the other hand, the application also provides an alcohol concentration information correction device, which comprises:

[0112] An alcohol concentration information acquisition module is configured to acquire alcohol concentration information detected by an electronic nose at a current time;

[0113] A correction model acquisition module is configured to acquire a correction model;

[0114] An alcohol concentration information correction module is configured to generate corrected alcohol concentration information according to the alcohol concentration information detected by the electronic nose at the current time and the correction model.

[0115] It should be noted that the method of one or more embodiments of the present application can be executed by a single device, such as a computer or a server, etc. The method of the present embodiment can also be applied to a distributed scenario, and completed by multiple devices cooperating with each other. In the case of such a distributed scenario, one of the multiple devices can only execute one or more steps in the method of one or more embodiments of the present application, and the multiple devices will interact with each other to complete the gearbox test bench lubricating oil temperature control method.

[0116] The above describes specific embodiments of the present application. Other embodiments are within the scope of the appended claims. In some cases, the actions or steps recited in the claims can be performed in a different order than the order in which they are recited in the embodiments and still achieve the desired results. In addition, the processes depicted in the figures do not necessarily require the particular order shown, or sequential order, to achieve the desired results. In certain implementations, multitasking and parallel processing can be advantageous.

[0117] Figure 3A more specific electronic device hardware structure schematic diagram provided by the embodiment is shown, and the device can include: a processor 1010, a memory 1020, an input / output interface 1030, a communication interface 1040, and a bus 1050. The processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040 are connected to each other through the bus 1050 for internal communication.

[0118] The processor 1010 can be implemented by a general-purpose CPU (Central Processing Unit), a microprocessor, an ASIC (Application Specific Integrated Circuit), or one or more integrated circuits, etc., for executing related programs to implement the alcohol concentration information correction method provided by the embodiments of the present specification.

[0119] The memory 1020 can be implemented by a ROM (Read Only Memory), a RAM (Random Access Memory), a static storage device, a dynamic storage device, etc. The memory 1020 can store an operating system and other application programs, and when the technical solutions provided by the embodiments of the present specification are implemented by software or firmware, the related program codes are stored in the memory 1020 and called and executed by the processor 1010.

[0120] The input / output interface 1030 is used to connect input / output modules to realize information input and output. The input / output modules can be configured as components in the device (not shown in the figure) or externally connected to the device to provide corresponding functions. The input device can include a keyboard, a mouse, a touch screen, a microphone, various sensors, etc., and the output device can include a display, a speaker, a vibrator, an indicator light, etc.

[0121] The communication interface 1040 is used to connect a communication module (not shown in the figure) to realize the communication interaction between the device and other devices. The communication module can realize communication through a wired manner (such as USB, network cable, etc.) or through a wireless manner (such as mobile network, WIFI, Bluetooth, etc.).

[0122] The bus 1050 includes a channel for transmitting information between various components (such as the processor 1010, the memory 1020, the input / output interface 1030, and the communication interface 1040) of the device.

[0123] It should be noted that although the above device only shows the processor 1010, the memory 1020, the input / output interface 1030, the communication interface 1040 and the bus 1050, in the specific implementation process, the device can also include other components necessary for normal operation. In addition, those skilled in the art can understand that the above device can also only contain components necessary to implement the embodiments of the present application, and does not necessarily contain all the components shown in the figure.

[0124] One embodiment of the present application provides a computer readable storage medium, which stores a computer program, and the computer program can implement the above alcohol concentration information correction method when executed by a processor.

[0125] The computer readable medium of the present embodiment includes permanent and non-permanent, removable and non-removable media, which can be realized by any method or technology to store information. The information can be computer readable instructions, data structures, program modules or other data. Examples of computer storage media include, but are not limited to, phase change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, compact disc read-only memory (CD-ROM), digital versatile disc (DVD) or other optical storage, magnetic cassette, magnetic tape magnetic disk storage or other magnetic storage devices, or any other non-transmission medium that can be used to store information that can be accessed by a computing device.

[0126] Those skilled in the art should understand that the above discussion of any embodiment is only exemplary and is not intended to imply that the scope of the present disclosure (including claims) is limited to these examples; under the idea of the present disclosure, the above embodiments or technical features in different embodiments can also be combined, the steps can be implemented in any order, and there are many other changes of different aspects of one or more embodiments of the present application as described above. In order to be brief, they are not provided in detail.

[0127] Additionally, to simplify the description and discussion, and so as not to obscure one or more embodiments of the description, well-known power supply / ground connections to integrated circuit (IC) chips and other components can or can not be shown in the provided figures. Furthermore, devices can be shown in block diagram form in order to avoid obscuring one or more embodiments of the description, and this also acknowledges the fact that the details in regard to the implementation of such block device are highly dependent on the platform within which the one or more embodiments of the description are being implemented (i.e., such details should be well within the purview of one of ordinary skill in the art). Where specific details are set forth in order to describe an illustrative embodiment of the disclosure, it will be apparent to one of ordinary skill in the art that the one or more embodiments of the description can be practiced without, or with variation of, these specific details. Thus, the description is to be considered as illustrative and not restrictive, and the scope of the one or more embodiments of the description is to be determined not with the assistance of the foregoing description alone, but rather in light of the appended claims in conjunction with recognizing the one or more embodiments of the description can over come a variety of non-anticipated

[0128] While the present disclosure has been described in connection with specific embodiments thereof, many alternatives, modifications and variations will be apparent to those skilled in the art in light of the foregoing description. For example, other memory architectures (e.g., dynamic RAM (DRAM)) can use the embodiments discussed.

[0129] The one or more embodiments of the description are intended to cover all such alternatives, modifications and variations as can come within the scope of the appended claims. Accordingly, any and all such modifications, variations or equivalents that fall within the scope of the one or more embodiments of the description should be considered within the scope of the disclosure.

Claims

1. A method of correcting alcohol concentration information, characterized by, The alcohol concentration information correction method comprises: obtaining alcohol concentration information detected by an electronic nose at a current time; obtaining a correction model; generating corrected alcohol concentration information according to the alcohol concentration information detected by the electronic nose at the current time and the correction model; The alcohol concentration information correction method further comprises: using a support vector machine as a classifier, respectively establishing a single-temperature model and two mixed-temperature models, and comparing the models with the correction model to prove that the correction model effectively reduces the influence of low temperature on the electronic nose detecting different concentrations of alcohol gas while reducing the sample requirement; Specifically, The single-temperature model is established by selecting data at T1 temperature as a training set, denoted as K1, and using a support vector machine classifier to test the prediction results at different temperature samples, i.e., data at T2 and T3 are used as a test set to classify the alcohol concentration in the gas; The two mixed-temperature models are established as follows: The alcohol mixed gas data of C1 and C2 at T1 and T3 temperatures are used as a mixed temperature data set, a support vector machine is used as a classifier to establish a first mixed temperature model, and the classification ability of the model for different alcohol concentrations of mixed gas at different temperatures is tested; wherein the data of T1 and T3 are used as a training set, denoted as K2; the data of T1, T2 and T3 are used as a test set, and the proportion of T1 and T3 in the training set and the test set is 7:3; The alcohol mixed gas data of C1 and C2 at T1, T2 and T3 temperatures are used as a mixed temperature data set, a support vector machine is used as a classifier to establish a second mixed temperature model, and the classification ability of the model for different alcohol concentrations of mixed gas at different temperatures is tested; wherein the data of T1, T2 and T3 are used as a training set, denoted as K3; the data of T1, T2 and T3 are used as a test set, and the proportion of T1, T2 and T3 in the training set and the test set is 7:3; Wherein, a support vector machine SVM is used as the classifier, and the correction model is an ICA temperature correction model; Wherein, the alcohol concentration information correction method further comprises: generating a correction model before the correction model is obtained; Wherein, the generation of the correction model comprises: obtaining multiple alcohol concentration sample information collected by the electronic nose in each preset environment; processing all alcohol concentration sample information using an ICA model to obtain a correction model; Wherein, the obtaining of the multiple alcohol concentration sample information collected by the electronic nose in each preset environment comprises: obtaining multiple alcohol concentration sample information collected by the electronic nose in each preset temperature and preset alcohol concentration space; Wherein, the obtaining of the multiple alcohol concentration sample information collected by the electronic nose in each preset environment comprises the following steps: The alcohol concentration in the simulated driver's drunk driving exhaled gas is 0.1 mg / L, containing 5% carbon dioxide, 16% oxygen, and the rest is nitrogen, denoted as C1; the alcohol concentration in the simulated driver's drunk driving exhaled gas is 0.5 mg / L, containing 5% carbon dioxide, 16% oxygen, and the rest is nitrogen, denoted as C2; The temperature gradient is divided into three parts: T1 = 20 ± 2℃, T2 = -10 ± 2℃, and T3 = -20 ± 2℃. The electronic nose was placed in a laboratory constant temperature chamber to adjust the ambient temperature, so that the temperature of the electronic nose was reduced to the target temperature. When the temperature reached the target temperature, two mixed gases, C1 and C2, with different alcohol concentrations, were released respectively. The electronic nose was used to collect gas information. Forty sets of samples of each concentration were measured at the same temperature. The smell data detected by the electronic nose under the three temperature conditions of T1, T2 and T3 of C1 and C2 is divided into 6 smell sample sets according to temperature and concentration, and is respectively recorded as L C1T1 、 C1T2 、 C1T3 、 C2T1 、 C2T2 、 C2T3 ; The step of using the ICA model to process all alcohol concentration sample information to obtain the correction model includes: Obtain feature information for each alcohol concentration sample; The characteristic information of all alcohol concentration samples was analyzed using the ICA model to obtain the independent components of the ICA model. A corrected model is obtained by training the independent components of the obtained ICA model using an identification method. The feature information for obtaining information on each alcohol concentration sample includes: The alcohol concentration sample information for each sample is standardized. Denoising was performed on the alcohol concentration sample information after each standardization process. Based on the alcohol concentration sample information after each noise reduction process, obtain the feature information corresponding to that alcohol concentration sample information; The standardization process for each alcohol concentration sample includes the following steps: The raw data obtained from the gas sensor array in the electronic nose were normalized to 0-1. The standardized formula is: ;in, It is the first of the original vectors. One element, It is the minimum value in the original vector. It is the maximum value in the original vector. It is the first of the generated vector data One element; According to each de-noising processing alcohol concentration sample information acquisition the alcohol concentration sample information corresponding feature information, extraction formula is: ; each data after de-noising is recorded as ; Specifically, the feature information of all alcohol concentration samples is analyzed using the ICA model to obtain the independent components of the ICA model, including: Construct a feature matrix from the feature information of all alcohol concentration samples; Perform mean removal processing on the feature matrix; Whiten the feature matrix after the mean removal process; Independent components of the ICA model are obtained from the feature matrix after mean removal and the whitening result. The feature matrix formed by all odor samples is denoted as F; The process of removing the mean from the feature matrix includes the following steps: The de-meaning formula is: ; wherein, is the mean value of ; The whitening of the mean-removed feature matrix includes the following steps: The whitening formula is: ; wherein, is a whitening matrix; The process of obtaining the independent components of the ICA model based on the mean-removed feature matrix and the whitening result includes the following steps: According to , the unmixing matrix is solved; wherein A is an unknown mixing matrix, and S is an unknown m (m≤N) dimensional signal source; obtaining independent components ; wherein are ICA components that are obtained by removing components strongly correlated with the ambient temperature disturbance using an Independent Component Analysis (ICA) model; by removing ICA output components strongly correlated with the ambient temperature disturbance, ICA components related to the true odor signal are obtained; The method of training the independent components of the obtained ICA model using recognition methods to obtain the correction model includes: A corrected model is obtained by training the independent components of the ICA model using a support vector machine; Among them, the selected ICA component U is trained and classified using the pattern recognition algorithm support vector machine to establish a correction model.

2. The alcohol concentration information correction method as described in claim 1, characterized in that, The noise reduction process for each standardized alcohol concentration sample includes: A 10Hz FFT filter was used to denoise each standardized alcohol concentration sample.

3. An alcohol concentration information correction device characterized by comprising: Used to perform the alcohol concentration information correction method according to any one of claims 1-2; The alcohol concentration information correction device includes: The alcohol concentration information acquisition module is used to acquire the alcohol concentration information detected by the electronic nose at the current moment; The correction model acquisition module is used to acquire the correction model; An alcohol concentration information correction module is configured to generate corrected alcohol concentration information based on alcohol concentration information detected by the electronic nose at the current time and the correction model.

Images