Intelligent alcohol detection method and system based on sensor fusion

Abstract

The invention discloses an intelligent alcohol detection method and system based on sensor fusion, and relates to the technical field of intelligent alcohol detection safety, and the method comprises the following steps: collecting alcohol concentration data, and processing the alcohol concentration data by adopting a self-adaptive filtering algorithm; transmitting the optimized alcohol concentration data to a detection cloud; and sharing alcohol concentration data detected by multiple users to finish supervision of an alcohol detection result. According to the invention, high-precision and low-noise alcohol concentration detection is realized, the intelligent level of the system is improved, the accuracy of data is ensured, and the overall efficiency and management efficiency of the system are improved, so that the alcohol detection system is remarkably improved in the aspects of precision, stability and data management, and the requirements of modern society on efficient and intelligent alcohol detection are met.

Description

Technical Field

[0001] The present invention relates to the field of intelligent alcohol detection safety technology, and in particular to an intelligent alcohol detection method and system based on sensor fusion. Background Art

[0002] The current society has increasingly higher requirements for alcohol testing of special groups such as drivers and workers to ensure safe production and public safety. However, most traditional alcohol detection methods rely on a single sensor, which is susceptible to environmental interference and sensor accuracy limitations, resulting in inaccuracy and instability in the test results. Based on the sensor fusion technology of the present invention, combined with the deep learning model and adaptive filtering algorithm, alcohol concentration data can be collected and processed more accurately, thereby greatly improving the accuracy and reliability of the test results.

[0003] At present, the existing technology has defects in the alcohol detection process, such as low detection accuracy, weak anti-interference ability, and difficulty in data sharing and management. The multi-sensor fusion, adaptive filtering and intelligent analysis model of the present invention can effectively solve these problems, realize multi-user collaborative management and data sharing, and ensure the efficiency of the detection process and the reliability of the detection results. Summary of the invention

[0004] In view of the problems existing in the existing intelligent alcohol detection method and system based on sensor fusion, the present invention is proposed.

[0005] Therefore, in order to solve the defects of the prior art in the alcohol detection process, such as low detection accuracy, weak anti-interference ability and difficulty in data sharing management, the present invention adopts multi-sensor fusion technology, deep learning model and adaptive filtering algorithm.

[0006] In order to solve the above technical problems, the present invention provides the following technical solutions:

[0007] In a first aspect, an embodiment of the present invention provides an intelligent alcohol detection method based on sensor fusion, which includes using the sensor fusion technology in combination with a deep learning model to collect alcohol concentration data, and using an adaptive filtering algorithm to process the alcohol concentration data;

[0008] Use intelligent analysis models to optimize the processed alcohol concentration data, and transmit the optimized alcohol concentration data to the detection cloud;

[0009] Build a multi-user collaborative management environment, share the alcohol concentration data detected by multiple users, and complete the supervision of alcohol test results.

[0010] As a preferred solution of the intelligent alcohol detection method based on sensor fusion described in the present invention, the sensor fusion technology includes establishing N sensors based on the intelligent detection system, decomposing each sensor, and determining the optimal solution based on the decomposition result S i (i=1, 2, ..., N), providing independent alcohol concentration reading A for each sensor i , the combined deep learning model includes converting the alcohol concentration reading A i As the input of the sensor, the weight vector of the sensor is calculated as the output. The specific calculation formula is:

[0011] w i =DNN(A,H)

[0012] Where A=[A1,A2,...,A N ] represents the current sensor reading, H represents the historical detection data matrix, and w i represents the weight vector of the sensor, DNN9A, H) represents the dynamic update of the neural network model parameters;

[0013] The output sensor weight vector is trained using the dynamic adjustment mechanism of the deep learning model. The specific calculation formula of the deep learning model is:

[0014]

[0015] Among them, A fused Indicates fusion, w i Represents the weight vector of the deep learning model, A i Indicates the alcohol concentration reading, and N indicates the sensor.

[0016] As a preferred solution of the intelligent alcohol detection method based on sensor fusion of the present invention, wherein: the collecting of alcohol concentration data includes determining the reading of the sensor before collecting;

[0017] If |A i -A median |>θ, that is, the reading of the i-th sensor is A i The median of all sensor readings A median When the absolute difference between exceeds the threshold θ,

[0018] Then i =0 and∑ j≠i w j =1, continue data collection, where j represents the index of the sensor;

[0019] If |A i -A median When |≤θ, the weight vector is calculated twice according to the deep learning model At this time, the sensor status is abnormal and data collection is stopped.

[0020] As a preferred solution of the intelligent alcohol detection method based on sensor fusion described in the present invention, the processing of alcohol concentration data includes decomposing the historical detection data matrix, the decomposition includes a sensor state estimation matrix and a sensor state observation matrix, and the specific calculation formula of the sensor state estimation matrix is:

[0021] x k+1 =Fx k +Bu k +w k

[0022] Among them, x k+1 represents the state variable at the next moment, F represents the state transfer matrix, x k represents the state variable at the current moment, B represents the control matrix, u k represents the control input at the current moment, w k represents the process noise of the sensor state estimation matrix;

[0023] The specific calculation formula of the sensor state observation matrix is:

[0024]

[0025] Among them, z k represents the observed variable at the current moment, H0 represents the fixed part of the observation matrix, and H j represents the dynamic part that depends on multi-dimensional data, A kj represents the reading of the jth sensor at the current moment, x k Represents the state variable at the current moment, v kj represents the sensor noise of the sensor state observation matrix.

[0026] As a preferred solution of the intelligent alcohol detection method based on sensor fusion of the present invention, wherein: the optimization of the processed alcohol concentration data includes adjusting the matrix using Gaussian distribution;

[0027] The sensor state estimation matrix is ​​adjusted so that the process noise of the sensor state estimation matrix satisfies the Gaussian distribution, and the specific formula is:

[0028] w k ~N(0,Q k )

[0029] Among them, w k represents the process noise of the sensor state estimation matrix, N represents the sensor, Q kAn adaptive algorithm for representing the sensor state estimation matrix;

[0030] The specific calculation formula of the adaptive algorithm of the sensor state estimation matrix is:

[0031]

[0032] Among them, Q k represents the adaptive algorithm of the sensor state estimation matrix, Q0 represents the covariance matrix of the fixed part of the sensor state estimation matrix, A i represents the reading of the i-th sensor, and N represents the sensor;

[0033] The sensor state observation matrix is ​​adjusted so that the sensor noise of the sensor state observation matrix satisfies the Gaussian distribution, and the specific formula is:

[0034] v kj ~N(0,R k )

[0035] Among them, v kj represents the sensor noise of the sensor state observation matrix, R k Adaptive algorithm for representing sensor state observation matrix;

[0036] The specific calculation formula of the adaptive algorithm of the sensor state observation matrix is:

[0037]

[0038] Among them, R k The adaptive algorithm represents the sensor state observation matrix, A kj represents the reading of the jth sensor at the current moment, R0 represents the observation noise covariance matrix of the fixed part of the sensor state observation matrix, R j represents the observation noise covariance matrix associated with the jth sensor.

[0039] As a preferred solution of the intelligent alcohol detection method based on sensor fusion described in the present invention, wherein: the transmission of the optimized alcohol concentration data to the detection cloud includes determining whether the sensor state estimation matrix is ​​adjusted;

[0040] If ||w k ||>σ w or||v k ||>σ v When the noise level changes, the noise covariance matrix Q is adjusted. k and the sensor state observation matrix R k ;

[0041] If ||w k||≤σ w or||v k ||≤σ v When , the noise covariance matrix Q is maintained k and the sensor state observation matrix R k constant.

[0042] As a preferred solution of the intelligent alcohol detection method based on sensor fusion described in the present invention, wherein: the sharing of alcohol concentration data detected by multiple users includes converting the adjusted matrix into alcohol transmission data A' ij Implement sharing, use the transmission protocol to transmit the alcohol transmission data during the alcohol transmission data process, store and manage the received alcohol transmission data on the central server, and use the multivariate linear regression model to analyze the alcohol transmission data;

[0043] If there are abnormal test results If When the data is stored in the central server, it is managed normally.

[0044] When a user performs M tests within a time period T, each test uses N sensors. The alcohol concentration dataset of all test results of the user within the time period T is {A' ijm};

[0045] At this time, the average alcohol concentration is calculated according to the alcohol concentration data set, and the specific calculation formula is:

[0046]

[0047] Among them, A' ijm It represents the alcohol concentration data processed by the i-th sensor during the user's M-th test. N represents the number of sensors used in each test, and M represents the number of tests performed within the time period T. Indicates the average alcohol concentration.

[0048] In a second aspect, an embodiment of the present invention provides an intelligent alcohol detection system based on sensor fusion, which includes: a processing module, which uses the sensor fusion technology and an adaptive filtering algorithm to process alcohol concentration data;

[0049] The optimization module uses an intelligent analysis model to optimize the processed alcohol concentration data and transmits the optimized alcohol concentration data to the detection cloud;

[0050] The detection module builds a multi-user collaborative management environment and shares the alcohol concentration data detected by multiple users in real time.

[0051] In a third aspect, an embodiment of the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the processor executes the computer program, any step of the above-mentioned intelligent alcohol detection method based on sensor fusion is implemented.

[0052] In a fourth aspect, an embodiment of the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program is executed by a processor, any step of the above-mentioned intelligent alcohol detection method based on sensor fusion is implemented.

[0053] The beneficial effects of the present invention are as follows: the present invention realizes high-precision, low-noise alcohol concentration detection through multi-sensor fusion technology, deep learning model and adaptive filtering algorithm. Sensor fusion technology improves the reliability of data acquisition, the deep learning model dynamically adjusts the sensor weight, improves the intelligence level of the system, and the adaptive filtering algorithm effectively removes noise and interference to ensure the accuracy of the data. The multi-user collaborative management environment enables the detection data to be shared and supervised in real time, improving the overall effectiveness and management efficiency of the system. The comprehensive application of these technical means has significantly improved the accuracy, stability and data management of the alcohol detection system, meeting the needs of modern society for efficient and intelligent alcohol detection. BRIEF DESCRIPTION OF THE DRAWINGS

[0054] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the following briefly introduces the drawings required for describing the embodiments. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without creative work. Among them:

[0055] Figure 1 A specific flow chart of an intelligent alcohol detection method and system based on sensor fusion provided by one embodiment of the present invention.

[0056] Figure 2 A specific flow chart of alcohol detection of an intelligent alcohol detection method and system based on sensor fusion provided by an embodiment of the present invention. DETAILED DESCRIPTION

[0057] In order to make the above-mentioned purposes, features and advantages of the present invention more obvious and easy to understand, the specific implementation methods of the present invention are described in detail below in conjunction with the drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, but not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary persons in the art without creative work should fall within the scope of protection of the present invention.

[0058] In the following description, many specific details are set forth to facilitate a full understanding of the present invention, but the present invention may also be implemented in other ways different from those described herein, and those skilled in the art may make similar generalizations without violating the connotation of the present invention. Therefore, the present invention is not limited to the specific embodiments disclosed below.

[0059] Secondly, the term "one embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The term "in one embodiment" that appears in different places in this specification does not necessarily refer to the same embodiment, nor does it refer to a separate or selective embodiment that is mutually exclusive with other embodiments.

[0060] The present invention is described in detail with reference to schematic diagrams. When describing the embodiments of the present invention, for the sake of convenience, the cross-sectional diagrams showing the device structure will not be partially enlarged according to the general scale, and the schematic diagrams are only examples, which should not limit the scope of protection of the present invention. In addition, in actual production, the three-dimensional dimensions of length, width and depth should be included.

[0061] At the same time, in the description of the present invention, it should be noted that the directions or positional relationships indicated by the terms "upper, lower, inner and outer" are based on the directions or positional relationships shown in the drawings, and are only for the convenience of describing the present invention and simplifying the description, rather than indicating or implying that the device or element referred to must have a specific direction, be constructed and operated in a specific direction, and therefore cannot be understood as limiting the present invention. In addition, the terms "first, second or third" are only used for descriptive purposes and cannot be understood as indicating or implying relative importance.

[0062] In the present invention, unless otherwise clearly specified and limited, the terms "install, connect, connect" should be understood in a broad sense, for example: it can be a fixed connection, a detachable connection or an integral connection; it can also be a mechanical connection, an electrical connection or a direct connection, or it can be indirectly connected through an intermediate medium, or it can be the internal communication of two components. For ordinary technicians in this field, the specific meanings of the above terms in the present invention can be understood according to specific circumstances.

[0063] Example 1

[0064] Reference Figure 1 and Figure 2 , which is the first embodiment of the present invention, and provides an intelligent alcohol detection method based on sensor fusion, comprising:

[0065] S1: Use sensor fusion technology combined with deep learning models to collect alcohol concentration data, and use adaptive filtering algorithms to process the alcohol concentration data.

[0066] Among them, the sensor fusion technology includes establishing N sensors based on the intelligent detection system, decomposing each sensor, and i (i=1, 2, ..., N), providing independent alcohol concentration reading A for each sensor i , combined with a deep learning model including the alcohol concentration reading A i As the input of the sensor, the weight vector of the sensor is calculated as the output. The specific calculation formula is:

[0067] w i =DNN(A,H)

[0068] Where A=[A1,A2,...,A N ] represents the current sensor reading, H represents the historical detection data matrix, and w i represents the weight vector of the sensor, and DNN(A, H) represents the dynamic update of the neural network model parameters;

[0069] The output sensor weight vector is trained using the dynamic adjustment mechanism of the deep learning model. The specific calculation formula of the deep learning model is:

[0070]

[0071] Among them, A fused Indicates fusion, w i Represents the weight vector of the deep learning model, A i Indicates the alcohol concentration reading, and N indicates the sensor.

[0072] S1.1: Collecting alcohol concentration data includes determining the sensor reading before collecting;

[0073] If |A i -A median |>θ, that is, the reading of the i-th sensor is A i The median of all sensor readings A median When the absolute difference between exceeds the threshold θ,

[0074] Then i =0 and∑ j≠i w j =1, continue data collection, where j represents the index of the sensor;

[0075] If |A i -A median When |≤θ, the weight vector is calculated twice according to the deep learning model At this time, the sensor status is abnormal and data collection is stopped.

[0076] Furthermore, the present invention utilizes a combination of sensor fusion technology and a deep learning model to collect alcohol concentration data, and uses an adaptive filtering algorithm to process the alcohol concentration data. First, N sensors are established through an intelligent detection system, and each sensor is decomposed to provide independent alcohol concentration readings. Then, these readings are input into a deep learning model to calculate the weight vector of the sensor, and the sensor weight vector is trained using the dynamic adjustment mechanism of the deep learning model. Specifically, the sensor reading and the historical detection data matrix work together to output the sensor weight vector. When collecting alcohol concentration data, the sensor reading is judged. If the absolute difference between a certain sensor reading and the median of all sensor readings exceeds a threshold, the sensor weight vector is adjusted; if the condition is not met, the sensor weight vector is calculated again according to the deep learning model. When an abnormal sensor state is detected, data collection is stopped. This method ensures high-precision, low-noise alcohol concentration detection, and significantly improves the accuracy and reliability of detection data.

[0077] S1.2: Processing the alcohol concentration data includes decomposing the historical detection data matrix, which includes the sensor state estimation matrix and the sensor state observation matrix. The specific calculation formula of the sensor state estimation matrix is:

[0078] x k+1 =Fx k +Bu k +w k

[0079] Among them, x k+1 represents the state variable at the next moment, F represents the state transfer matrix, x k represents the state variable at the current moment, B represents the control matrix, u k represents the control input at the current moment, w k represents the process noise of the sensor state estimation matrix;

[0080] The specific calculation formula of the sensor state observation matrix is:

[0081]

[0082] Among them, z k represents the observed variable at the current moment, H0 represents the fixed part of the observation matrix, and H j represents the dynamic part that depends on multi-dimensional data, A kj represents the reading of the jth sensor at the current moment, x k Represents the state variable at the current moment, v kj represents the sensor noise of the sensor state observation matrix.

[0083] S2: Use the intelligent analysis model to optimize the processed alcohol concentration data, and transmit the optimized alcohol concentration data to the detection cloud.

[0084] Among them, optimizing the processed alcohol concentration data includes adjusting the matrix using Gaussian distribution;

[0085] The sensor state estimation matrix is ​​adjusted so that the process noise of the sensor state estimation matrix satisfies the Gaussian distribution. The specific formula is:

[0086] w k ~N(0,Q k )

[0087] Among them, w k represents the process noise of the sensor state estimation matrix, N represents the sensor, Q k An adaptive algorithm for representing the sensor state estimation matrix;

[0088] The specific calculation formula of the adaptive algorithm of the sensor state estimation matrix is:

[0089]

[0090] Among them, Q k represents the adaptive algorithm of the sensor state estimation matrix, Q0 represents the covariance matrix of the fixed part of the sensor state estimation matrix, A i represents the reading of the i-th sensor, and N represents the sensor;

[0091] The sensor state observation matrix is ​​adjusted so that the sensor noise of the sensor state observation matrix satisfies the Gaussian distribution. The specific formula is:

[0092] v kj ~N(0,R k )

[0093] Among them, v kj represents the sensor noise of the sensor state observation matrix, R k Adaptive algorithm for representing sensor state observation matrix;

[0094] The specific calculation formula of the adaptive algorithm of the sensor state observation matrix is:

[0095]

[0096] Among them, R k The adaptive algorithm represents the sensor state observation matrix, A kj represents the reading of the jth sensor at the current moment, R0 represents the observation noise covariance matrix of the fixed part of the sensor state observation matrix, R jrepresents the observation noise covariance matrix associated with the jth sensor.

[0097] Furthermore, the processed alcohol concentration data is optimized using an intelligent analysis model, and the optimized data is transmitted to the detection cloud. The optimization process includes adjusting the sensor state estimation matrix and the sensor state observation matrix using Gaussian distribution. For the sensor state estimation matrix, the process noise is adjusted by Gaussian distribution to satisfy the Gaussian distribution characteristics. Assuming that the mean of the process noise is 0 and the variance is 0.01, the noise parameters of the sensor state estimation matrix can be adjusted so that its process noise satisfies these characteristics. For example, if the process noise of a sensor at a certain moment is 0.02, after adjustment, its mean can be close to 0 and its variance can be 0.01. The specific adaptive algorithm is implemented by adjusting the covariance matrix of the fixed part and the sensor reading. For example, when the covariance matrix of the fixed part is 0.001 and the reading is 0.5, the new noise parameter can be 0.0015 through algorithm adjustment. The sensor state observation matrix is ​​also similarly adjusted to make its sensor noise satisfy the Gaussian distribution. Assuming that the mean of the observation noise is 0 and the variance is 0.005, the noise parameters can be adjusted to meet these statistical characteristics. For example, if the reading of a sensor at the current moment is 0.45, the adjusted observation noise parameter is 0.0048. Through these adjustments, the optimized alcohol concentration data is more accurate and stable. The optimized data is transmitted to the detection cloud to ensure real-time and efficient supervision and management. The data adjustment parameters are shown in Table 1 below:

[0098] Table 1 Data adjustment parameter table

[0099]

[0100] This table shows the various parameters and their corresponding values ​​used to optimize the alcohol concentration data, including the noise adjustment coefficients of the sensor state estimation matrix and the sensor state observation matrix in different states, the sensor weight vector, the historical data matrix, and the fixed part covariance matrix and the observation noise covariance matrix related to the adaptive algorithm. These parameters are adjusted through the Gaussian distribution model to ensure that the sensor noise meets the expected distribution requirements, thereby optimizing the transmission and analysis of alcohol concentration data in the detection cloud.

[0101] S2.1: Transmitting the optimized alcohol concentration data to the detection cloud includes adjusting the sensor state estimation matrix;

[0102] If ||w k ||>σ w or||v k ||>σ v When the noise level changes, the noise covariance matrix Q is adjusted. kand the sensor state observation matrix R k ;

[0103] If ||w k ||≤σ w or||v k ||≤σ v When , the noise covariance matrix Q is maintained k and the sensor state observation matrix R k constant.

[0104] S3: Build a multi-user collaborative management environment to share alcohol concentration data detected by multiple users in real time and complete the supervision of alcohol test results.

[0105] The sharing of alcohol concentration data detected by multiple users includes converting the adjusted matrix into alcohol transmission data A' ij Implement sharing, use the transmission protocol to transmit the alcohol transmission data during the alcohol transmission data process, store and manage the received alcohol transmission data on the central server, and use the multivariate linear regression model to analyze the alcohol transmission data;

[0106] If there are abnormal test results If When the data is stored in the central server, it is managed normally.

[0107] When a user performs M tests within a time period T, each test uses N sensors. The alcohol concentration dataset of all test results of the user within the time period T is {A' ijm};

[0108] At this time, the average alcohol concentration is calculated based on the alcohol concentration data set. The specific calculation formula is:

[0109]

[0110] Among them, A' ijm It represents the alcohol concentration data processed by the i-th sensor during the user's M-th test. N represents the number of sensors used in each test, and M represents the number of tests performed within the time period T. Indicates the average alcohol concentration.

[0111] Furthermore, when building a multi-user collaborative management environment, the alcohol concentration data detected by each user is shared and supervised. First, the adjusted matrix of each user's alcohol concentration data is converted into a transmittable data format, and a special transmission protocol is used for data transmission. For example, suppose a user performs 5 tests in one day, each test uses 3 sensors, and the alcohol concentration readings obtained are 0.05, 0.07, 0.06, 0.08, 0.09, 0.07, 0.04, 0.06, 0.05, 0.07, 0.08, 0.06, 0.05, 0.07, 0.04. Through the transmission protocol, these data are sent to the central server for storage and management.

[0112] On the central server, a multivariate linear regression model is used to analyze the received alcohol concentration data. If the test results are abnormal, such as the average alcohol concentration of a certain test exceeds the preset safety threshold, an alarm mechanism is triggered. For example, if the average alcohol concentration of a certain test exceeds 0.08, an alarm is triggered.

[0113] For normal test data, routine storage management is performed on the central server. When analyzing these data, the average alcohol concentration of each user within a certain period of time will be calculated. For example, if a user performs 15 tests in a week using 3 sensors, and the test data are 0.05, 0.06, 0.07, 0.04, 0.05, 0.06, 0.03, 0.04, 0.05, etc., a total of 45 data points, then the average alcohol concentration of these 15 tests will be calculated. In this way, collaborative management, sharing and supervision of multi-user test data can be achieved, thereby improving the accuracy and real-time performance of alcohol detection.

[0114] In a preferred embodiment, a smart alcohol detection system based on sensor fusion includes a processing module that uses sensor fusion technology and an adaptive filtering algorithm to process alcohol concentration data;

[0115] The optimization module uses an intelligent analysis model to optimize the processed alcohol concentration data and transmits the optimized alcohol concentration data to the detection cloud;

[0116] The detection module builds a multi-user collaborative management environment and shares the alcohol concentration data detected by multiple users in real time.

[0117] The above-mentioned unit modules may be embedded in or independent of a processor in a computer device in the form of hardware, or may be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to the above-mentioned modules.

[0118] The computer device may be a terminal, and the computer device includes a processor, a memory, a communication interface, a display screen and an input device connected via a system bus. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be achieved through WIFI, an operator network, NFC (near field communication) or other technologies. The display screen of the computer device may be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device may be a touch layer covering the display screen, or a key, trackball or touchpad provided on the housing of the computer device, or an external keyboard, touchpad or mouse, etc.

[0119] In summary, the present invention realizes high-precision, low-noise alcohol concentration detection through multi-sensor fusion technology, deep learning model and adaptive filtering algorithm. Sensor fusion technology improves the reliability of data acquisition, the deep learning model dynamically adjusts the sensor weight, improves the intelligence level of the system, and the adaptive filtering algorithm effectively removes noise and interference to ensure the accuracy of the data. The multi-user collaborative management environment enables the detection data to be shared and supervised in real time, improving the overall effectiveness and management efficiency of the system. The comprehensive application of these technical means has significantly improved the accuracy, stability and data management of the alcohol detection system, meeting the needs of modern society for efficient and intelligent alcohol detection.

[0120] Example 2

[0121] Reference Figure 1 and Figure 2 , which is the second embodiment of the present invention, and this embodiment provides an intelligent alcohol detection method based on sensor fusion. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through simulation experiments.

[0122] Ten volunteers were selected, and each volunteer underwent alcohol tests three times a day for 30 days. Each test used five different types of sensors, including infrared fiber optic sensors, gas sensors and semiconductor sensors. Each sensor independently recorded alcohol concentration data. In order to simulate different alcohol intakes, the volunteers drank different amounts of alcoholic beverages before each test.

[0123] During the data collection process, the data recorded by each sensor is transmitted to the system. First, the system uses an adaptive filtering algorithm to perform preliminary processing on the data to eliminate noise and outliers. For example, in a test on a certain day, a volunteer's sensor recorded the following data: 0.05, 0.06, 0.07, 0.08, and 0.09. The outlier value of a certain sensor was detected and filtered, and the processed data was further optimized. The weight vector of the sensor was calculated through a deep learning model, and the data was adjusted according to the Gaussian distribution to make it more accurate.

[0124] The system transmits the processed and optimized alcohol concentration data to the detection cloud and stores it on the central server. During this process, a specific transmission protocol is used to ensure the integrity and security of the data. Through the central server, the system uses a multivariate linear regression model to analyze all collected data. For example, in a 30-day period, each volunteer underwent a total of 90 tests, with a total of 4,500 data points.

[0125] The experimental results show that through a multi-user collaborative management environment, the alcohol test data of all volunteers are effectively shared and supervised. For abnormal results detected, such as the average alcohol concentration of a volunteer exceeding 0.08 on a certain day, the system successfully triggered the alarm mechanism. For normal test data, the system performed routine storage and analysis. The accuracy and reliability of the method were verified by calculating the average alcohol concentration of each volunteer within 30 days.

[0126] Finally, the experiment shows that the intelligent alcohol detection method based on sensor fusion can accurately and real-time detect and share the alcohol concentration data of multiple users, and effectively realize the supervision and management of alcohol detection results. The alcohol detection experimental data is shown in Table 2 below:

[0127] Table 2 Alcohol detection experimental data table

[0128]

[0129] Through this simplified experimental data table, we can clearly see the total number of tests for each volunteer within 30 days, the number of sensors used, the initial data sample, the number of abnormal tests, and the final average alcohol concentration. The experimental data show that this technical solution can effectively detect and share alcohol concentration data and accurately monitor alcohol test results. The comparison between the present invention and the prior art is shown in Table 3 below:

[0130] Table 3 Comparison between the present invention and the prior art

[0131]

[0132] The above table compares the differences between the existing technical solutions and the technical solutions of the present invention in many aspects, and focuses on demonstrating the advantages of the present invention. By adopting sensor fusion technology and adaptive filtering algorithms based on intelligent detection systems, combined with deep learning models for data processing, the present invention has obvious advantages in data accuracy, flexibility and adaptability. In addition, the present invention constructs a multi-user collaborative management environment, realizes data sharing and secure transmission, and uses a multivariate linear regression model for data analysis, which improves the intelligence level of the system. Experimental verification shows that the present invention has significant improvements in processing effects and system security, fully demonstrating its superiority in alcohol detection applications.

[0133] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention rather than to limit it. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solutions of the present invention may be modified or replaced by equivalents without departing from the spirit and scope of the technical solutions of the present invention, which should all be included in the scope of the claims of the present invention.

Claims

1. An intelligent alcohol detection method based on sensor fusion, characterized in that: include, The sensor fusion technology is used in combination with a deep learning model to collect alcohol concentration data, and an adaptive filtering algorithm is used to process the alcohol concentration data; Use intelligent analysis models to optimize the processed alcohol concentration data, and transmit the optimized alcohol concentration data to the detection cloud; Build a multi-user collaborative management environment, share the alcohol concentration data detected by multiple users, and complete the supervision of alcohol test results.

2. The intelligent alcohol detection method based on sensor fusion as claimed in claim 1, characterized in that: The sensor fusion technology includes establishing N sensors based on an intelligent detection system, decomposing each sensor, and i (i=1, 2, ..., N), providing independent alcohol concentration reading A for each sensor i , the combined deep learning model includes converting the alcohol concentration reading A i As the input of the sensor, the weight vector of the sensor is calculated as the output. The specific calculation formula is: w i =DNN(A,H) Where A=[A1,A2,...,A N ] represents the current sensor reading, H represents the historical detection data matrix, and w i represents the weight vector of the sensor, and DNN(A, H) represents the dynamic update of the neural network model parameters; The output sensor weight vector is trained using the dynamic adjustment mechanism of the deep learning model. The specific calculation formula of the deep learning model is: Among them, A fused Indicates fusion, w i Represents the weight vector of the deep learning model, A i Indicates the alcohol concentration reading, and N indicates the sensor.

3. The intelligent alcohol detection method based on sensor fusion as claimed in claim 2, characterized in that: The collecting of alcohol concentration data includes determining the reading of the sensor before collecting; If |A i -A median |>θ, that is, the reading of the i-th sensor is A i The median of all sensor readings A median When the absolute difference between exceeds the threshold θ, Then i =0 and∑ j≠i w j =1, continue data collection, where j represents the index of the sensor; If |A i -A median When |≤θ, the weight vector is calculated twice according to the deep learning model At this time, the sensor status is abnormal and data collection is stopped.

4. The intelligent alcohol detection method based on sensor fusion as claimed in claim 3, characterized in that: The processing of the alcohol concentration data includes decomposing the historical detection data matrix, wherein the decomposition includes a sensor state estimation matrix and a sensor state observation matrix. The specific calculation formula of the sensor state estimation matrix is: x k+1 =Fx k +This k +w k Among them, x k+1 represents the state variable at the next moment, F represents the state transfer matrix, x k represents the state variable at the current moment, B represents the control matrix, u k represents the control input at the current moment, w k represents the process noise of the sensor state estimation matrix; The specific calculation formula of the sensor state observation matrix is: Among them, z k represents the observed variable at the current moment, H0 represents the fixed part of the observation matrix, and H j represents the dynamic part that depends on multi-dimensional data, A kj represents the reading of the jth sensor at the current moment, x k Represents the state variable at the current moment, v kj represents the sensor noise of the sensor state observation matrix.

5. The intelligent alcohol detection method based on sensor fusion as claimed in claim 4, characterized in that: The optimization of the processed alcohol concentration data includes adjusting the matrix using Gaussian distribution; The sensor state estimation matrix is ​​adjusted so that the process noise of the sensor state estimation matrix satisfies the Gaussian distribution, and the specific formula is: w k ~N(0,Q k ) Among them, w k represents the process noise of the sensor state estimation matrix, N represents the sensor, Q k An adaptive algorithm for representing the sensor state estimation matrix; The specific calculation formula of the adaptive algorithm of the sensor state estimation matrix is: Among them, Q k represents the adaptive algorithm of the sensor state estimation matrix, Q0 represents the covariance matrix of the fixed part of the sensor state estimation matrix, A i represents the reading of the i-th sensor, and N represents the sensor; The sensor state observation matrix is ​​adjusted so that the sensor noise of the sensor state observation matrix satisfies the Gaussian distribution, and the specific formula is: v kj ~N(0,R k ) Among them, v kj represents the sensor noise of the sensor state observation matrix, R k Adaptive algorithm for representing sensor state observation matrix; The specific calculation formula of the adaptive algorithm of the sensor state observation matrix is: Among them, R k The adaptive algorithm represents the sensor state observation matrix, A kj represents the reading of the jth sensor at the current moment, R0 represents the observation noise covariance matrix of the fixed part of the sensor state observation matrix, R j represents the observation noise covariance matrix associated with the jth sensor.

6. The intelligent alcohol detection method based on sensor fusion as claimed in claim 5, characterized in that: The transmitting of the optimized alcohol concentration data to the detection cloud includes determining the sensor state estimation matrix and the sensor state observation matrix, and adjusting them; If ||w k ||>σ w or||v k ||>σ v When the noise level changes, the noise covariance matrix Q is adjusted. k and the sensor state observation matrix R k ; If ||w k ||≤σ w or||v k ||≤σ v When , the noise covariance matrix Q is maintained k and the sensor state observation matrix R k constant.

7. The intelligent alcohol detection method based on sensor fusion as claimed in claim 6, characterized in that: The sharing of alcohol concentration data detected by multiple users includes converting the adjusted matrix into alcohol transmission data A' ij Implement sharing, use the transmission protocol to transmit the alcohol transmission data during the alcohol transmission data process, store and manage the received alcohol transmission data on the central server, and use the multivariate linear regression model to analyze the alcohol transmission data; If there are abnormal test results If When the data is stored in the central server, it is managed normally. When a user performs M tests within a time period T, each test uses N sensors. The alcohol concentration dataset of all test results of the user within the time period T is {A' ijm }; At this time, the average alcohol concentration is calculated according to the alcohol concentration data set, and the specific calculation formula is: Among them, A' ijm It represents the alcohol concentration data processed by the i-th sensor during the user's M-th test. N represents the number of sensors used in each test, and M represents the number of tests performed within the time period T. Indicates the average alcohol concentration.

8. An intelligent alcohol detection system based on sensor fusion, based on the intelligent alcohol detection method based on sensor fusion according to any one of claims 1 to 7, characterized in that: include, A processing module, which uses the sensor fusion technology and an adaptive filtering algorithm to process the alcohol concentration data; The optimization module uses an intelligent analysis model to optimize the processed alcohol concentration data and transmits the optimized alcohol concentration data to the detection cloud; The detection module builds a multi-user collaborative management environment and shares the alcohol concentration data detected by multiple users in real time.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the intelligent alcohol detection method based on sensor fusion according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the intelligent alcohol detection method based on sensor fusion according to any one of claims 1 to 7 are implemented.

Images