Intelligent waistcoat-assisted environment analysis early warning method and system

Abstract

The invention discloses an intelligent waistcoat-assisted environment analysis early warning method and system, and relates to the technical field of environment monitoring, and the method comprises the steps: collecting and preprocessing multi-source environment data, and carrying out the feature extraction and fusion of the preprocessed environment data; a deep neural network DNN model is constructed, the risk value of the environmental data is evaluated, and the risk value of the environmental data is monitored in real time and early warning is carried out; and constructing a visual interface to display the risk value and early warning of the environmental data in real time, and storing, collecting and analyzing the generated environmental data. According to the method, multi-source environment data is collected and preprocessed, and feature extraction and fusion are performed on the preprocessed environment data; a deep neural network DNN model is constructed, the risk value of the environmental data is evaluated, and the risk value of the environmental data is monitored in real time and early warning is carried out; the accuracy and effectiveness of environmental data analysis are improved, and the flexibility and safety of environmental risk early warning are enhanced.

Description

Technical Field

[0001] The present invention relates to the technical field of environmental monitoring, and in particular to an intelligent vest-assisted environmental analysis and early warning method and system. Background Art

[0002] With the acceleration of modern urbanization, environmental pollution is becoming increasingly serious, especially in some industrialized cities. The monitoring and analysis of environmental factors such as air quality, temperature, humidity, and harmful gases have become crucial. Existing environmental monitoring technologies mainly rely on fixed monitoring sites, which have limited coverage and are not very practical and personal. In order to make up for this shortcoming, wearable devices such as smart vests have gradually been used in the field of environmental monitoring in recent years. These devices integrate multiple sensors to collect real-time data on the surrounding environment while the user is moving, and combine big data analysis and artificial intelligence technology to achieve dynamic environmental early warning. Existing technologies still have a lot of room for optimization in the accuracy and efficiency of multi-source data fusion, feature extraction, and intelligent early warning.

[0003] The deficiencies of existing technologies in environmental monitoring and early warning are mainly reflected in the following aspects: traditional environmental monitoring methods usually rely on a single type of data source, while ignoring the diversity of environmental data and the importance of multi-source data fusion. This single data processing method makes it impossible to comprehensively and accurately assess risks in complex environments. Most existing environmental early warning models are based on static threshold settings and cannot flexibly respond to environmental changes under different times and conditions, which can easily lead to false alarms or missed reports. Summary of the invention

[0004] In view of the problems existing in the above-mentioned existing intelligent vest-assisted environmental analysis and early warning methods, the present invention is proposed.

[0005] Therefore, the problem to be solved by the present invention is that traditional environmental monitoring methods usually rely on a single type of data source, while ignoring the diversity of environmental data and the importance of multi-source data fusion. This single data processing method makes it impossible to comprehensively and accurately assess risks in complex environments. Most of the existing environmental early warning models are based on static threshold settings and cannot flexibly respond to environmental changes at different times and conditions, which can easily cause false alarms or missed reports.

[0006] To solve the above technical problems, the present invention provides the following technical solutions: an intelligent vest-assisted environmental analysis and early warning method, which includes collecting and preprocessing multi-source environmental data, extracting and fusing features of the preprocessed environmental data; constructing a deep neural network DNN model to evaluate the risk value of environmental data, and monitor the risk value and early warning of environmental data in real time; constructing a visual interface to display the risk value and early warning of environmental data in real time, and storing, collecting and analyzing the generated environmental data.

[0007] As a preferred solution of the smart vest-assisted environmental analysis and early warning method of the present invention, wherein: the collection of multi-source environmental data and pre-processing refers to installing an air temperature sensor in the chest area of ​​the smart vest, installing a humidity sensor in the waist area of ​​the smart vest, installing an air quality sensor in the central position of the back of the smart vest, installing a harmful gas sensor in the chest of the smart vest near the breathing area, and installing a radiation sensor in the shoulder area of ​​the smart vest, and the sensors collect multi-source environmental data at a unified collection frequency; The multi-source environmental data includes temperature, humidity, air quality, harmful gas concentration and radiation intensity data; The interquartile range method (IQR) is used to detect anomalies in the collected multi-source environmental data and delete outliers. The weighted moving average filtering algorithm is used to denoise the collected multi-source environmental data, and the denoised multi-source environmental data is standardized.

[0008] As a preferred solution of the intelligent vest-assisted environmental analysis and early warning method of the present invention, wherein: the feature extraction and fusion of the pre-processed environmental data refers to extracting the main features from the multi-source environmental data using principal component analysis PCA ; Use logarithmic function to extract main features Perform logarithmic transformation to obtain the eigenvalue after logarithmic transformation ; Use the Gaussian kernel function to calculate the similarity weight of the i-th type of environmental data at time t ; Use the time decay function to calculate the time decay weight of the i-th type of environmental data at time t ; Use a nonlinear scoring mechanism to process the environmental data nonlinearly and calculate the score of the environmental data at time t , the formula is: , where m is the number of categories of environmental data, is the standardized value of the j-th type of environmental data at time t; Based on the eigenvalues ​​after logarithmic transformation , calculate the similarity weight of the i-th type of environmental data at time t , the time attenuation weight of the i-th type of environmental data at time t And the score of the environmental data at time t , construct the feature fusion formula, and fuse the extracted main features. The formula is: ,in is the feature fusion result at time t, T and 0 are the upper and lower limits of the integration, k is the number of main features extracted by principal component analysis PCA, is the ith regularization parameter.

[0009] As a preferred solution of the intelligent vest-assisted environmental analysis and early warning method of the present invention, wherein: the construction of a deep neural network DNN model to evaluate the risk value of environmental data refers to collecting environmental training data from a development data platform and performing preprocessing, feature extraction and fusion to generate a training set; Build a deep neural network DNN model, including input layer, hidden layer and output layer; Set the format of the input layer to fusion features ; Use the training set to train the deep neural network DNN model, and use the loss function and Adam optimizer to iteratively optimize the model parameters; The fusion features Bring it into the trained deep neural network DNN model to obtain the risk value of the environmental data at time t .

[0010] As a preferred solution of the intelligent vest-assisted environmental analysis and early warning method of the present invention, the risk value of the real-time monitoring environmental data and the early warning index construct an adaptive dynamic threshold, and set the early warning threshold, the formula is: ,in is the warning threshold at time t, u is the adjustment coefficient, is the standard deviation of the risk value; The risk value of the environmental data at time t The warning threshold at time t For comparison, if When the area is identified as a dangerous area, the smart vest will sound an audible and visual alarm and and warning thresholds The frequency of collecting multi-source environmental data is adjusted by the ratio of ; like When the area is detected, it is judged as a normal area, and the smart vest continues to monitor the multi-source environmental data at the frequency of initial collection of multi-source environmental data.

[0011] As a preferred solution of the intelligent vest-assisted environmental analysis and early warning method described in the present invention, wherein: the construction of a visualization interface to display the risk value and early warning of environmental data in real time refers to using the D3.js front-end data visualization framework to develop a visualization interface, using data visualization tools to display the risk value of environmental data, and performing a review operation on verified user requests.

[0012] As a preferred solution of the environmental analysis and early warning method assisted by the smart vest described in the present invention, wherein: the storage of environmental data collected and analyzed refers to storing the risk values ​​of the collected multi-source environmental data and the environmental data generated by analysis in a database, the database is sorted in chronological order and marked with corresponding tags, the collected multi-source environmental data and the environmental data generated by analysis are synchronously backed up in the cloud, and the backup data is regularly checked for integrity.

[0013] Another object of the present invention is to provide an intelligent vest-assisted environmental analysis and early warning system, which comprises: The collection and fusion module is used to collect and preprocess multi-source environmental data, extract features from the preprocessed environmental data, and fuse them; Build an early warning module to build a deep neural network (DNN) model, evaluate the risk value of environmental data, and monitor the risk value and early warning of environmental data in real time; The visualization storage module is used to build a visualization interface to display the risk value and warning of environmental data in real time, and to store, collect and analyze the environmental data generated.

[0014] A computer device comprises: a memory and a processor; the memory stores a computer program, and the processor implements the steps of the above-mentioned intelligent vest-assisted environmental analysis and early warning method when executing the computer program.

[0015] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the steps of the above-mentioned intelligent vest-assisted environmental analysis and early warning method.

[0016] The beneficial effects of the present invention are as follows: the present invention collects multi-source environmental data and performs preprocessing, extracts features from the preprocessed environmental data and performs fusion; constructs a deep neural network (DNN) model to evaluate the risk value of environmental data, and monitors the risk value and early warning of environmental data in real time; improves the accuracy and effectiveness of environmental data analysis, and enhances the flexibility and security of environmental risk early warning. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings required for use in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other accompanying drawings can be obtained based on these accompanying drawings without paying creative work.

[0018] Figure 1 Schematic diagram of the process of environmental analysis and early warning method assisted by smart vest.

[0019] Figure 2This is a structural diagram of the environmental analysis and early warning system assisted by the smart vest. DETAILED DESCRIPTION

[0020] In order to make the above-mentioned objects, features and advantages of the present invention more obvious and easy to understand, the specific embodiments of the present invention are described in detail below in conjunction with the accompanying drawings.

[0021] 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.

[0022] 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 is it a separate or selective embodiment that is mutually exclusive with other embodiments.

[0023] Example 1, reference Figure 1 , which is the first embodiment of the present invention, and which provides an intelligent vest-assisted environmental analysis and early warning method, the intelligent vest-assisted environmental analysis and early warning method comprising: S1, collect and preprocess multi-source environmental data, extract features from the preprocessed environmental data and perform fusion; Specifically, collecting multi-source environmental data and preprocessing it means installing a temperature sensor on the chest area of ​​the smart vest, installing a humidity sensor on the waist area of ​​the smart vest, installing an air quality sensor on the central back of the smart vest, installing a harmful gas sensor on the chest of the smart vest near the breathing area, and installing a radiation sensor on the shoulder area of ​​the smart vest, and the sensors collect multi-source environmental data at a unified collection frequency; The multi-source environmental data includes temperature, humidity, air quality, harmful gas concentration and radiation intensity data; The interquartile range method (IQR) is used to detect anomalies in the collected multi-source environmental data and delete outliers. The weighted moving average filtering algorithm is used to denoise the collected multi-source environmental data, and the denoised multi-source environmental data is standardized.

[0024] Temperature, humidity, air quality, harmful gas concentration and radiation intensity data are independent and complementary. Only by collecting these multi-source environmental data at the same time can the smart vest provide a comprehensive and accurate environmental risk assessment. Each data type plays a unique role in specific health risk assessments. By using the IQR method, abnormal data can be detected more robustly and misjudgments caused by environmental fluctuations can be reduced, thereby improving data reliability and the accuracy of multi-source environmental data, laying a solid foundation for subsequent data processing and analysis. The multidimensional characteristics of environmental data make denoising processing complicated, but the weighted moving average filter can process noise from multiple data sources at the same time, retain real and effective environmental change information, enhance the system's real-time response capability to environmental changes, and improve the accuracy of environmental risk assessment. Through standardization, all data are converted to the same scale, so that the importance of each type of data can be fairly reflected in subsequent data fusion and analysis, ensuring that the fusion process of different environmental data is more balanced, avoiding data imbalance problems, and improving the overall accuracy of environmental analysis and early warning.

[0025] Furthermore, feature extraction and fusion of the preprocessed environmental data refers to extracting the main features from multi-source environmental data using principal component analysis (PCA). ; Use logarithmic function to extract main features Perform logarithmic transformation to obtain the eigenvalue after logarithmic transformation , the formula is: ,in Extract the main features for the i-th time at time t, is a very small positive number; Use the Gaussian kernel function to calculate the similarity weight of the i-th type of environmental data at time t , the formula is: ,in is the standardized value of the i-th type of environmental data at time t, is the bandwidth parameter of the kernel function, which is used to control the range of similarity; Use the time decay function to calculate the time decay weight of the i-th type of environmental data at time t , the formula is: ,in is the time decay rate parameter, which is used to control the decay rate of each main feature over time; In order to extract the main features of the i-th time t from the multi-source environmental data through principal component analysis PCA, these environmental data are dynamic and time-related. Air quality, temperature, etc. are not constant. They will change over time. The nonlinear scoring mechanism is used to process the environmental data nonlinearly and calculate the score of the environmental data at time t. , the formula is: , where m is the number of categories of environmental data, is the standardized value of the j-th type of environmental data at time t; Environmental data may change dramatically in some cases, such as a sudden increase in temperature or concentration of harmful gases. These changes usually mean potential risks. In order to ensure that the system is more sensitive to these abnormal data, these abnormal values ​​are amplified through nonlinearity, so that the system can capture potential environmental risks more quickly. If linear weighting or mean calculation is used, the system will respond slowly to these sudden changes, which may lead to missing important risk signals. In environmental data, small fluctuations such as slight changes in humidity should not be overreacted in many cases. The logarithmic square term smoothes slight fluctuations to avoid the system being too sensitive to these small fluctuations, thereby reducing false alarms or overreactions. If a linear method is used to handle these slight fluctuations, the system will frequently respond to these unnecessary changes, resulting in an increase in false alarms. Environmental data is multidimensional, and the data collected by each sensor may have an impact on the overall risk assessment. These multidimensional data must be uniformly scored for further fusion and processing. If the multidimensional data is not scored, but relies solely on a certain sensor or data source, the system's response will be limited to a single dimension and cannot accurately assess the overall risk. Directly processing single-dimensional data or using linear weighting will ignore the interaction of multiple dimensions.

[0026] Based on the eigenvalues ​​after logarithmic transformation , calculate the similarity weight of the i-th type of environmental data at time t , the time attenuation weight of the i-th type of environmental data at time t And the score of the environmental data at time t , construct the feature fusion formula, and fuse the extracted main features. The formula is: ,in is the feature fusion result at time t, T and 0 are the upper and lower limits of the integration, k is the number of main features extracted by principal component analysis PCA, is the i-th main feature, is the standardized value of the i-th type of environmental data at time t, is the bandwidth parameter of the kernel function, is the i-th main feature The square of the Euclidean distance between the normalized value of the i-th category environmental data at time t, is the i-th regularization parameter, is the time decay rate parameter, m is the number of categories of environmental data, is the standardized value of the j-th type of environmental data at time t.

[0027] The relationship between environmental data is usually not linear, and the traditional weighted average method cannot effectively capture these nonlinear relationships. By introducing the RBF kernel function, the nonlinear similarity between the principal component and the real-time data can be quantified. This method is necessary for feature fusion. The kernel function can automatically adjust the contribution of the feature according to the similarity of the data. It is an effective tool for capturing complex multidimensional data features. Existing technologies mostly use linear weighted methods for data fusion, which easily ignores the nonlinear correlation between data and cannot accurately evaluate the interaction between multidimensional data. The fixed weight method is difficult to adapt to variable environmental data. The logarithmic function is used to perform nonlinear processing on the eigenvalues ​​to ensure that the eigenvalues ​​will not generate negative values ​​during the fusion process, and it amplifies smaller features and suppresses excessive eigenvalues. For the features in the environmental data, some data have a large variation range, and directly using these features will lead to imbalance. Logarithmic function processing is necessary because it can adjust the distribution of eigenvalues ​​so that the contribution of all features to the fusion process is more balanced, especially small values ​​will not be ignored. , and large values ​​will not dominate the entire process. Traditional assessment methods directly use the original eigenvalues ​​for linear weighting, resulting in features with larger eigenvalues ​​dominating the results, while smaller features have insufficient influence. Through the logarithmic function, smaller eigenvalues ​​are amplified and larger eigenvalues ​​are compressed to ensure that each feature can play a reasonable role when fused. The Gaussian kernel function captures the complex relationship between data in a nonlinear way, making the fusion process of data features more flexible and adaptive, thereby improving the accuracy of feature fusion. Integrating the data of the entire time period can capture the long-term trend of environmental data, rather than just assessing the risk at a single moment. Dynamic time integration can provide more comprehensive risk assessment results and is therefore irreplaceable. Existing environmental risk assessment models are mostly static assessments that only consider the data at the current moment and ignore the dynamic impact of historical data. By introducing the time decay function and time integration, the model can automatically adjust over time according to the impact of historical data, thereby more flexibly assessing risks and improving the dynamic response capability of the model.

[0028] Extracting the main features through PCA can effectively reduce the dimension of the data and retain the most representative information in the data, thereby improving computing efficiency and reducing redundancy. The use of PCA not only simplifies the data processing process, but also solves the problem of how to extract useful information from multi-source data, enabling the system to respond to environmental changes more quickly, effectively improving data processing efficiency, and enhancing the real-time and stability of the system. The Gaussian kernel function calculates the distance between data points and gives similar data a higher weight, thereby improving the accuracy of feature fusion. Compared with the traditional linear weight calculation method, it solves the problem that data association is difficult to accurately quantify in a nonlinear environment, making data fusion more accurate and feature-rich. The effect is particularly significant when processing multi-source data with complex internal relationships. Traditional data processing methods usually ignore the impact of time on data, resulting in the system's over-reliance on historical data and slow response. The introduction of the time decay function solves this problem, enabling the system to respond quickly based on the latest data, greatly improving the accuracy and timeliness of real-time warnings. By comprehensively considering the timeliness, similarity and normalized scores of the data, various types of data are effectively integrated, which not only solves the imbalance problem in the multi-source data fusion process, but also provides high-quality feature data for subsequent risk assessment and warning, thereby improving the accuracy and robustness of the entire environmental analysis system.

[0029] S2. Build a deep neural network (DNN) model to evaluate the risk value of environmental data, monitor the risk value of environmental data in real time and issue early warnings; Specifically, building a deep neural network DNN model to assess the risk value of environmental data refers to collecting environmental training data from the development data platform and performing preprocessing, feature extraction and fusion to generate a training set; Build a deep neural network DNN model, including input layer, hidden layer and output layer; Set the format of the input layer to fusion features ; Use the training set to train the deep neural network DNN model, and use the loss function and Adam optimizer to iteratively optimize the model parameters; The fusion features Bring it into the trained deep neural network DNN model to obtain the risk value of the environmental data at time t .

[0030] Different from traditional environmental monitoring methods, the preprocessing step of the present invention not only improves the quality of data, but also solves the problem of difficult integration of multi-dimensional data through multi-source data fusion, ensures high-quality data input, lays a solid foundation for subsequent DNN model training, and improves the accuracy and robustness of risk assessment. By introducing the DNN model, the system can better capture the complex relationships in environmental data and make more accurate predictions of environmental risks. Through the multi-layer network structure, the system can extract deep feature information from complex environmental data, significantly improving the accuracy and sensitivity of risk assessment. The Adam optimizer can adaptively adjust the learning rate of the model to reduce Reduce training time and improve convergence speed. By using the loss function, the model can continuously adjust parameters in each iteration so that the prediction results gradually approach the true value. Compared with the traditional gradient descent algorithm, the Adam optimizer has better robustness, especially suitable for complex and noisy environmental data. While ensuring the accuracy of the model, it greatly improves the training efficiency, avoids the model from falling into local optimality, and enhances the system's adaptability to complex environmental data. Through dynamic data input and deep learning models, it can more accurately assess the potential risks of the environment. By introducing dynamic environmental data, the system can analyze the risk level of the current environment in real time, enhancing the accuracy and real-time nature of environmental warnings.

[0031] Furthermore, the risk value and warning index of real-time monitoring of environmental data are used to construct an adaptive dynamic threshold and set the warning threshold. The formula is: ,in is the warning threshold at time t, u is the adjustment coefficient, is the standard deviation of the risk value; The risk value of the environmental data at time t The warning threshold at time t For comparison, if When the area is identified as a dangerous area, the smart vest will sound an audible and visual alarm and and warning thresholds The frequency of collecting multi-source environmental data is adjusted by the ratio of , the formula is: ,in The frequency of initial collection of multi-source environmental data, is the frequency boost coefficient, which is used to control the sensitivity of the acquisition frequency boost; like When the area is detected, it is judged as a normal area, and the smart vest continues to monitor the multi-source environmental data at the frequency of initial collection of multi-source environmental data.

[0032] Fusion Features The result is a numerical value. The square root function is mathematically valid and is usually used to weaken the strength of a feature or to indicate a change of a certain magnitude in a physical sense. By smoothing the feature fusion result, the excessive impact of high-risk data on the threshold can be reduced, and the system can be prevented from frequent false alarms. The excessive fluctuation of the threshold setting when the environmental data changes dramatically is reduced, thereby improving the stability and robustness of the system. The existing threshold setting method based on the mean and standard deviation cannot comprehensively consider the complex multidimensional characteristics and real-time changes of the environment, which leads to a lag in threshold adjustment and an inability to quickly respond to environmental changes. The present invention introduces a nonlinear adjustment mechanism to adjust the threshold to ensure that the system can smoothly handle the dramatic changes of complex data and avoid overreaction. The threshold setting in existing technologies usually only considers environmental data of a single dimension, or fuses data through a simple weighted method, which cannot fully reflect the interactions in a multidimensional environment. Feature fusion combines multiple environmental data dimensions to ensure that the system can fully integrate multidimensional feature data and accurately reflect the overall changes in the current environment. The existing dynamic threshold method based on standard deviation relies entirely on data volatility and ignores the multidimensionality and complex characteristics of the data. Therefore, it is easy to cause threshold adjustment lags in complex environments. By using standard deviation as an auxiliary adjustment item and combining nonlinear scoring and multidimensional feature fusion, it is ensured that when the system responds to environmental data fluctuations, it not only considers data volatility, but also integrates multidimensional characteristics and makes flexible adjustments.

[0033] By introducing adaptive dynamic thresholds, the system can adjust the warning standards according to the fluctuations of environmental data. The threshold is dynamically adjusted as the environmental data changes, so that the system shows greater flexibility and accuracy when dealing with changing environments. Through the formula, the system can not only adjust the threshold according to the standard deviation, but also flexibly set warning strategies of different sensitivities through the adjustment coefficient u. Through dynamic adjustment, the warning system can quickly adapt to different environmental conditions, significantly reduce the false alarm rate, and improve the accuracy and effectiveness of the warning. The warning threshold is dynamically changing and is adjusted in real time according to data fluctuations and environmental conditions. When the environmental risk value exceeds the threshold, the system will automatically determine it as a dangerous area and issue an audible and visual alarm to remind the user. Compared with the existing system based on static thresholds, the present invention solves the problem of false alarms caused by environmental changes. By comparing the risk value and the warning threshold in real time, the system can make judgments in time and confirm It ensures a quick response when environmental risks increase, improving the accuracy and timeliness of early warnings. The introduction of the adjustment coefficient u effectively solves the problem that the early warning system cannot flexibly adapt to different scenarios. Through the adjustment of u, the system can strike a balance between sensitivity and stability, ensuring that the early warning system can work effectively in different scenarios. By adjusting the initial frequency and the frequency boost coefficient, the system can increase the frequency of data collection during high-risk periods, ensuring that the system can quickly collect more environmental data when the environment changes more drastically, thereby improving the ability to perceive risks. In low-risk situations, maintaining the initial frequency can avoid excessive consumption of resources. Under normal circumstances, the system still maintains basic monitoring of the environment to ensure that potential risks are not missed due to too low a frequency. Compared with the fixed frequency design of the prior art, the invention adjusts the frequency in combination with the risk assessment results, making the system's monitoring more intelligent.

[0034] S3, build a visual interface to display the risk value and warning of environmental data in real time, and store the environmental data collected and analyzed; Specifically, building a visualization interface to display the risk value and warning of environmental data in real time means using the D3.js front-end data visualization framework to develop a visualization interface, using data visualization tools to display the risk value of environmental data, and performing review operations on verified user requests.

[0035] For the risk assessment of environmental data, D3.js can not only present traditional charts, but also help users understand the changing trends behind the data more deeply through its rich interactive functions. Compared with the existing static visualization solutions, the use of D3.js significantly improves users' sensitivity and insight into data, realizes real-time and dynamic display of environmental data, enhances the interactivity and visualization of the system, and enables users to grasp the environmental risk status more intuitively. When the environmental risk value exceeds the warning threshold, the system can prompt users of potential environmental crises through graphical highlighting or animation. Compared with traditional digital or tabular presentation, this display method helps users make decisions quickly in complex data, improves the system's response efficiency and user experience, and through identity authentication, not only the security of data is guaranteed, but also the compliance of the system is ensured, solving the problems of insufficient data security and access control in existing technologies. By strengthening data access control, the system is ensured to be sound in protecting data privacy and security, and the security of the system and the trust of users are improved.

[0036] Furthermore, storing the environmental data collected and generated by analysis means storing the risk values ​​of the collected multi-source environmental data and the environmental data generated by the analysis in a database, sorting the database in chronological order and marking the corresponding tags, synchronously backing up the collected multi-source environmental data and the environmental data generated by the analysis to the cloud, and regularly performing integrity checks on the backup data.

[0037] Through chronological sorting, it can not only help users quickly locate environmental data for a specific time period, but also provide a structured data foundation for subsequent trend analysis. This time sorting method helps the system to manage and query data more efficiently. By storing data in chronological order, the efficiency of data management is significantly improved, and a solid foundation is provided for subsequent data analysis and trend prediction. By synchronizing and backing up data to the cloud, the reliability and security of the data are greatly enhanced. Even if the local device fails, the data will not be lost. Cloud backup allows data to be accessed at any time on multiple platforms, meeting the needs of mobility and convenience. Through cloud backup, data redundancy is guaranteed, which greatly reduces the risk of data loss and improves data accessibility. Through regular integrity testing, all backup data is ensured to be intact. Through regular testing, the integrity of the backup data is ensured, and the stability and reliability of the system in long-term operation are improved.

[0038] Example 2, reference Figure 2 , which is the second embodiment of the present invention. This embodiment is different from the previous embodiment and provides an intelligent vest-assisted environmental analysis and early warning system. The collection and fusion module is used to collect and preprocess multi-source environmental data, extract features from the preprocessed environmental data, and fuse them; Build an early warning module to build a deep neural network (DNN) model, evaluate the risk value of environmental data, and monitor the risk value and early warning of environmental data in real time; The visualization storage module is used to build a visualization interface to display the risk value and warning of environmental data in real time, and to store, collect and analyze the environmental data generated.

[0039] If the functions are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or the part of the technical solution, can be embodied in the form of a software product. The computer software product is stored in a storage medium, including several instructions to enable a computer device (which can be a personal computer, server, or network device, etc.) to perform all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk, etc., which can store program codes.

[0040] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as an ordered list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in conjunction with such instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in conjunction with such instruction execution systems, devices or apparatuses.

[0041] More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection with one or more wires (electronic device), a portable computer disk case (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be a paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium, followed by editing, deciphering or, if necessary, processing in another suitable manner, and then stored in a computer memory.

[0042] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit with a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit with a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.

Claims

1. An intelligent vest-assisted environmental analysis and early warning method, characterized in that: include, Collect and preprocess multi-source environmental data, extract features from the preprocessed environmental data, and perform fusion; Build a deep neural network (DNN) model to evaluate the risk value of environmental data, and monitor the risk value and early warning of environmental data in real time; Build a visual interface to display the risk value and warning of environmental data in real time, and store the environmental data collected and analyzed.

2. The intelligent vest-assisted environmental analysis and early warning method according to claim 1, characterized in that: The collecting of multi-source environmental data and preprocessing thereof refers to installing an air temperature sensor on the chest area of ​​the smart vest, installing a humidity sensor on the waist area of ​​the smart vest, installing an air quality sensor on the central back of the smart vest, installing a harmful gas sensor on the chest of the smart vest near the breathing area, and installing a radiation sensor on the shoulder area of ​​the smart vest, and the sensors are used to collect multi-source environmental data at a unified collection frequency; The multi-source environmental data includes temperature, humidity, air quality, harmful gas concentration and radiation intensity data; The interquartile range method (IQR) is used to detect anomalies in the collected multi-source environmental data and delete outliers. The weighted moving average filtering algorithm is used to denoise the collected multi-source environmental data, and the denoised multi-source environmental data is standardized.

3. The intelligent vest-assisted environmental analysis and early warning method according to claim 2, characterized in that: The feature extraction and fusion of the pre-processed environmental data refers to extracting the main features from the multi-source environmental data using principal component analysis (PCA) ; Use logarithmic function to extract main features Perform logarithmic transformation to obtain the eigenvalue after logarithmic transformation ; Use the Gaussian kernel function to calculate the similarity weight of the i-th type of environmental data at time t ; Use the time decay function to calculate the time decay weight of the i-th type of environmental data at time t ; Use a nonlinear scoring mechanism to process the environmental data nonlinearly and calculate the score of the environmental data at time t , the formula is: , where m is the number of categories of environmental data, is the standardized value of the j-th type of environmental data at time t; Based on the eigenvalues ​​after logarithmic transformation , calculate the similarity weight of the i-th type of environmental data at time t , the time attenuation weight of the i-th type of environmental data at time t And the score of the environmental data at time t , construct the feature fusion formula, and fuse the extracted main features. The formula is: ,in is the feature fusion result at time t, T and 0 are the upper and lower limits of the integration, k is the number of main features extracted by principal component analysis PCA, is the ith regularization parameter.

4. The smart vest-assisted environmental analysis and early warning method according to claim 3, characterized in that: The construction of the deep neural network DNN model to evaluate the risk value of environmental data refers to collecting environmental training data from the development data platform and performing preprocessing, feature extraction and fusion to generate a training set; Build a deep neural network DNN model, including input layer, hidden layer and output layer; Set the format of the input layer to fusion features ; Use the training set to train the deep neural network DNN model, and use the loss function and Adam optimizer to iteratively optimize the model parameters; The fusion features Bring it into the trained deep neural network DNN model to obtain the risk value of the environmental data at time t .

5. The smart vest-assisted environmental analysis and early warning method according to claim 4, characterized in that: The risk value and early warning of real-time monitoring of environmental data refer to building an adaptive dynamic threshold and setting an early warning threshold. The formula is: ,in is the warning threshold at time t, u is the adjustment coefficient, is the standard deviation of the risk value; The risk value of the environmental data at time t The warning threshold at time t For comparison, if When the area is identified as a dangerous area, the smart vest will sound an audible and visual alarm and and warning thresholds The frequency of collecting multi-source environmental data is adjusted by the ratio of ; like When the area is detected, it is judged as a normal area, and the smart vest continues to monitor the multi-source environmental data at the frequency of initial collection of multi-source environmental data.

6. The smart vest-assisted environmental analysis and early warning method according to claim 5, characterized in that: The construction of a visualization interface to display the risk value and warning of environmental data in real time refers to developing a visualization interface using the D3.js front-end data visualization framework, using data visualization tools to display the risk value of environmental data, and performing a review operation on verified user requests.

7. The smart vest-assisted environmental analysis and early warning method according to claim 6, characterized in that: The storage of collected and analyzed environmental data refers to storing the risk values ​​of the collected multi-source environmental data and the environmental data generated by analysis in a database, sorting the database in chronological order and marking corresponding tags, synchronously backing up the collected multi-source environmental data and the environmental data generated by analysis to the cloud, and regularly performing integrity checks on the backup data.

8. An intelligent vest-assisted environmental analysis and early warning system based on the intelligent vest-assisted environmental analysis and early warning method according to any one of claims 1 to 7, characterized in that: include, The collection and fusion module is used to collect and preprocess multi-source environmental data, extract features from the preprocessed environmental data, and fuse them; Build an early warning module to build a deep neural network (DNN) model, evaluate the risk value of environmental data, and monitor the risk value and early warning of environmental data in real time; The visualization storage module is used to build a visualization interface to display the risk value and warning of environmental data in real time, and to store, collect and analyze the environmental data generated.

9. A computer device comprising: Memory and processor; The memory stores a computer program, characterized in that when the processor executes the computer program, the steps of the smart vest-assisted environmental analysis and early warning method described in 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 smart vest-assisted environmental analysis and early warning method described in any one of claims 1 to 7 are implemented.

Images