Chemical early warning interception method and device, electronic equipment and medium

Abstract

The embodiment of the invention discloses a chemical early warning interception method and device, electronic equipment and a medium. A specific embodiment of the method comprises the following steps: carrying out identity recognition on a target chemical detection terminal set to obtain a detection terminal identity recognition information set; performing state information extraction on the multi-modal target chemical information set to obtain a chemical multi-modal state information set; performing data deduplication on the multi-modal state information set of the chemicals and then performing association identification to obtain a multi-modal association information set; performing chemical risk identification on the de-duplicated multi-modal chemical information set to obtain a to-be-tracked chemical information set; and performing trajectory tracking early warning on the to-be-tracked chemical article information set to obtain a chemical article trajectory early warning information set, and performing early warning interception on the to-be-tracked chemical article information set. According to the embodiment, a data barrier can be broken, data sharing and safety are improved, and the accuracy of accurate interception and risk early warning of target chemicals is improved.

Description

TECHNICAL FIELD

[0001] Embodiments of the present disclosure relate to the technical field of computer technology, and specifically to a chemical substance early warning interception method and device, electronic equipment and medium. BACKGROUND

[0002] Dangerous chemical substances are an important influencing factor threatening public safety. Precise identification and early warning tracking and interception of dangerous chemical substances involve multiple departments, which need to cooperate to reduce the challenges of dangerous chemical substances to public safety. For early warning interception of chemical substances, the commonly used way is that multiple departments involved identify dangerous chemical substances according to the data related to chemical substances owned by themselves to obtain dangerous chemical substance identification information. Then, the dangerous chemical substance identification information is tracked and early warned for interception.

[0003] However, in practice, it is found that when the above method is used to early warn and intercept chemical substances, the following technical problems often exist: due to the data barriers of multiple departments involved, the data sharing ability is low and the data is difficult to effectively integrate, the chemical substance identification and early warning interception are performed through single data, which leads to low accuracy of chemical substance identification, a large number of missed and false detections of dangerous chemical substances, low accuracy of dangerous chemical substance interception, and low accuracy of risk early warning, increasing the risk caused by dangerous chemical substances and reducing social safety.

[0004] The above information disclosed in this BACKGROUND section is only for enhancing the understanding of the background of the present disclosure concept, and therefore, it can contain information that does not form the prior art known to those of ordinary skill in the art in the country. SUMMARY

[0005] The summary section of the present disclosure is used to introduce the concepts in a brief form, which will be described in detail in the specific embodiments section. The summary section of the present disclosure is not intended to identify key or essential features of the claimed technical solutions, nor is it intended to limit the scope of the claimed technical solutions.

[0006] Some embodiments of the present disclosure propose a chemical substance early warning interception method, device, electronic equipment and medium to solve one or more of the technical problems mentioned in the background section.

[0007] In a first aspect, some embodiments of this disclosure provide a chemical substance early warning and interception method, comprising: responding to the detection of a data connection request information set of a target chemical substance detection terminal set, performing identity recognition on the target chemical substance detection terminal set to obtain a detection terminal identity recognition information set; controlling at least one target chemical substance detection terminal represented by the detection terminal identity recognition information set to pass the recognition; extracting chemical substance status information from the multimodal target chemical substance information set corresponding to the at least one target chemical substance detection terminal to obtain a chemical substance multimodal status information set; and performing data deduplication on the chemical substance multimodal status information set. The process involves processing to obtain a deduplicated multimodal chemical information set; performing multimodal association identification on the deduplicated multimodal chemical information set to obtain a multimodal association information set; performing chemical risk identification on the deduplicated multimodal chemical information set based on the multimodal association information set to obtain a chemical information set to be tracked; performing trajectory tracking and early warning on the chemical information set to be tracked to obtain a chemical trajectory early warning information set; and controlling a chemical early warning device set and a chemical interception device set based on the chemical trajectory early warning information set to perform early warning and interception on the chemical information set to be tracked.

[0008] Secondly, some embodiments of this disclosure provide a chemical substance early warning and interception device, comprising: an identity recognition unit configured to, in response to a detected data connection request information set of a target chemical substance detection terminal set, perform identity recognition on the target chemical substance detection terminal set to obtain a detection terminal identity recognition information set; a chemical substance status information extraction unit configured to control the detection terminal identity recognition information set to represent at least one target chemical substance detection terminal that has passed the identification, and extract chemical substance status information from the multimodal target chemical substance information set corresponding to the at least one target chemical substance detection terminal to obtain a chemical substance multimodal status information set; and a data collision deduplication unit configured to perform data deduplication on the chemical substance multimodal status information set. The system processes and obtains a deduplicated multimodal chemical information set; a multimodal association identification unit is configured to perform multimodal association identification on the deduplicated multimodal chemical information set to obtain a multimodal association information set; a chemical risk identification unit is configured to perform chemical risk identification on the deduplicated multimodal chemical information set based on the multimodal association information set to obtain a chemical information set to be tracked; and an early warning and interception unit is configured to perform trajectory tracking and early warning on the chemical information set to be tracked to obtain a chemical trajectory early warning information set, and to control a chemical early warning device set and a chemical interception device set based on the chemical trajectory early warning information set to perform early warning and interception on the chemical information set to be tracked corresponding to the chemical information set to be tracked.

[0009] Thirdly, some embodiments of this disclosure provide an electronic device, including: one or more processors; and a storage device having one or more programs stored thereon, such that when the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any implementation of the first aspect.

[0010] Fourthly, some embodiments of this disclosure provide a computer-readable medium having a computer program stored thereon, wherein the computer program, when executed by a processor, implements the method as described in any implementation of the first aspect.

[0011] The various embodiments of this disclosure have the following beneficial effects: The chemical warning and interception methods of some embodiments of this disclosure, by integrating multimodal data from multiple data sources, can break down data barriers, improve data sharing and security, enhance the accurate interception of target chemicals, and reduce the accident rate caused by target chemicals. Specifically, the reason for the low accuracy of related chemical identification, with numerous missed and false detections, and the resulting low accuracy of chemical warning and interception, thus reducing social security, is that data barriers exist among multiple departments involved, data sharing capabilities are low, and data is difficult to integrate effectively. Using single data for chemical identification and warning interception leads to low accuracy in chemical identification, numerous missed and false detections, resulting in low accuracy in target chemical interception and risk warning, increasing the risk caused by target chemicals and reducing social security. Based on this, the chemical warning and interception methods of some embodiments of this disclosure can first, in response to the detected data connection request information set of the target chemical detection terminal set, perform identity recognition on the target chemical detection terminal set to obtain the detection terminal identity recognition information set. Here, identity recognition can accurately determine the identity information of each target chemical detection terminal, which can, to a certain extent, avoid the harm caused by unauthorized and forged data and improve data security. Secondly, by controlling the identity recognition information set of the aforementioned detection terminals to represent at least one target chemical detection terminal that has passed the identification, chemical status information is extracted from the multimodal target chemical information set corresponding to the aforementioned at least one target chemical detection terminal, resulting in a multimodal chemical status information set. Here, extracting status information from the data of different target chemical detection terminals not only improves data security but also reduces the waste of storage resources. Thirdly, the aforementioned multimodal chemical status information set undergoes data deduplication processing to obtain a deduplicated multimodal chemical information set. Here, conflicting and identical data in the multimodal chemical status information set can be removed, improving the data quality and reducing the data volume of the deduplicated multimodal chemical information set. Subsequently, multimodal association recognition is performed on the aforementioned deduplicated multimodal chemical information set to obtain a multimodal association information set. Here, data barriers between different data sources can be broken down, improving data sharing capabilities and data correlation performance, thereby enhancing the linkage performance and efficiency of different data sources during subsequent target chemical identification. Then, based on the aforementioned multimodal correlation information set, chemical risk identification is performed on the deduplicated multimodal chemical information set to obtain the information set of chemicals to be tracked. Here, using comprehensive and high-quality data for chemical risk identification can improve the accuracy of target chemical identification.Finally, trajectory tracking and early warning are performed on the aforementioned set of chemicals to be tracked, resulting in a chemical trajectory early warning information set. Based on this information set, a set of chemical early warning devices and a set of chemical interception devices are controlled to intercept the corresponding set of chemicals to be tracked. This improves the accuracy of early warnings and the precision of interception for target chemicals, reduces the probability of dangerous events caused by target chemicals, enhances social security, reduces the operational load on the chemical early warning and interception device sets, and lowers their failure rate. Therefore, this chemical early warning and interception system, by integrating multimodal data stored and detected by multiple target chemical detection terminals, can break down data barriers, improve data sharing and security, enhance the precision of target chemical interception and early warning, reduce the probability of dangerous events caused by target chemicals, and improve social security. Attached Figure Description

[0012] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and elements are not necessarily drawn to scale.

[0013] Figure 1 This is a flowchart of some embodiments of the chemical warning and interception method according to this disclosure; Figure 2 These are schematic diagrams illustrating the structure of some embodiments of the chemical warning and interception device according to this disclosure; Figure 3 This is a schematic diagram of the structure of an electronic device suitable for implementing some embodiments of the present disclosure. Detailed Implementation

[0014] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0015] It should also be noted that, for ease of description, only the parts relevant to the invention are shown in the accompanying drawings. Unless otherwise specified, the embodiments and features described in this disclosure can be combined with each other.

[0016] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0017] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0018] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0019] This disclosure will now be described in detail with reference to the accompanying drawings and embodiments.

[0020] Figure 1 A flow 100 of some embodiments of a chemical warning and interception method according to the present disclosure is shown. The chemical warning and interception method includes the following steps: Step 101: In response to the detection of the data connection request information set of the target chemical substance detection terminal set, the target chemical substance detection terminal set is identified to obtain the detection terminal identification information set.

[0021] In some embodiments, the executing entity (e.g., an electronic device) of the above-described chemical substance early warning and interception method can respond to a data connection request information set of a detected target chemical substance detection terminal set via a wired or wireless connection, and identify the target chemical substance detection terminal set to obtain a detection terminal identification information set. The target chemical substance detection terminals in the above-described target chemical substance detection terminal set can be data sources from different levels and different databases used for storing and detecting target chemical substance information. The target chemical substance information can be information about substances that pose a certain risk and endanger social security. For example, the target chemical substances can include, but are not limited to, at least one of the following: refined oil, natural gas, civilian explosives, flammable materials, and explosive materials. The target chemical substance detection terminal set can include, but is not limited to, at least one of the following: terminals storing data sources from enterprises that detect target chemical substances, terminals storing data sources from enterprises that sell target chemical substances, terminals storing data sources from transportation of target chemical substances, terminals storing data sources from provinces, cities, and other target chemical substance data sources. The target chemical substance detection terminal can be a server. The detection terminal identification information in the above-described detection terminal identification information set can be information used to characterize the identity of the target chemical substance detection terminal. For example, the aforementioned detection terminal identification information may include, but is not limited to, at least one of the following: detection terminal permission information, detection terminal device encoding information, detection terminal server IP address, and detection terminal level information.

[0022] In some optional implementations of certain embodiments, the process of identifying the target chemical detection terminal set in response to the detection of a data connection request information set of the target chemical detection terminal set to obtain a detection terminal identification information set may include the following steps: The first step involves extracting identity information from the aforementioned data connection request information set to obtain a terminal digital certificate information set, a terminal current dynamic token set, and a terminal request signature information set. Specifically, the terminal digital certificate information in the terminal digital certificate information set can be a digital certificate used to prove the legitimacy of the target chemical testing terminal. This terminal digital certificate information may include: a unique terminal identifier, a terminal public key, a signature of the terminal digital certificate content encrypted with a private key, and the digital certificate's validity period. The current dynamic token in the terminal current dynamic token set can be a one-time, real-time verified random password. This current dynamic token can be a token generated by a TOTP (Time-based One-Time Password) generator and TOTP verifier deployed on the target chemical testing terminal. The terminal request signature information in the terminal request signature information set can be digital signature information used to ensure that the data sent by the target chemical testing terminal is not tampered with. This terminal request signature information can be signature information obtained by the target chemical testing terminal using its own private key to sign the communication connection timestamp and the terminal current dynamic token.

[0023] The second step involves verifying the aforementioned terminal digital certificate information set to obtain a digital certificate verification result information set. This set includes verification information on the validity period of the terminal digital certificate, the legality of the signature encrypted with the private key, and the validity status of the terminal digital certificate. Signature verification can be performed by decrypting the signature using the terminal's public key to obtain a first hash value. The verifier then calculates a second hash value using the same hash algorithm. If the first and second hash values ​​match, the verification is successful; otherwise, it fails. The verifier can be the target chemical substance detection terminal receiving the connection request.

[0024] The third step is to select at least one target chemical detection terminal from the above target chemical detection terminal set that has passed the digital certificate verification result information verification, thus obtaining the selected target chemical detection terminal set.

[0025] The fourth step involves generating a set of target dynamic token groups to be verified based on the dynamic token key set and the current timestamp set included in the terminal digital certificate information set corresponding to the target chemical detection terminal set after the above screening. The target dynamic token groups to be verified in the above set can be tokens representing identity information predicted and calculated by the verifying party.

[0026] As an example, the aforementioned execution entity can perform the following input steps for each terminal digital certificate in the aforementioned terminal digital certificate information set: First, determine the difference between the current timestamp and the start time of the verifier's TOTP verifier, and its ratio to a preset time step, to obtain a timestamp ratio. The preset time step can be a pre-set fixed time interval. For example, the preset time step could be 30 seconds. Second, round down the timestamp ratio to obtain the current time step. Then, input the dynamic token key, the current time step, the current time step minus one current terminal dynamic token, and the current time step plus one current terminal dynamic token sequentially into the TOTP verifier to obtain the target dynamic token group to be verified.

[0027] The fifth step involves comparing the target set of dynamic tokens to be verified with the current set of dynamic tokens on the terminal to obtain a dynamic token comparison result information set. This dynamic token comparison result information indicates whether the terminal's current dynamic token is in the target set of dynamic tokens to be verified. If the terminal's current dynamic token is in the target set of dynamic tokens to be verified, the token verification is successful; otherwise, the token verification fails.

[0028] Step 6: Perform data tamper-proof verification on the aforementioned terminal request signature information set to obtain a data tamper-proof verification information set. The data tamper-proof verification information in this set can be used to verify whether the terminal request signature information has been tampered with. In practice, the executing entity can first decrypt the terminal request signature information set using the data source public key set to obtain a decrypted request information set. If each communication connection timestamp in the communication connection timestamp set included in the decrypted request information set is within the range of its corresponding current timestamp, and the dynamic token set is identical to the dynamic token corresponding to the dynamic token key set, then the terminal request signature information set verification is successful; otherwise, verification fails.

[0029] Step 7: Based on the dynamic token comparison result information set and the data anti-tampering verification information set mentioned above, generate the detection terminal identity recognition information set.

[0030] As an example, the aforementioned executing entity may, in response to determining that at least one dynamic token comparison result in the aforementioned dynamic token comparison result information has passed verification, also correspondingly indicates successful verification in the aforementioned data anti-tampering verification information set, determine the terminal unique identifier and the data range authorized by the target chemical detection terminal as the terminal identification information set. In response to determining that at least one dynamic token comparison result in the aforementioned dynamic token comparison result information set has passed verification, but correspondingly indicates failed verification in the aforementioned data anti-tampering verification information set, determine the failed verification as the detection terminal identification information set.

[0031] Step 102: Control the identification information of the detection terminal to represent at least one target chemical detection terminal that has passed the identification, and extract the chemical status information from the multimodal target chemical information set corresponding to at least one target chemical detection terminal to obtain the multimodal chemical status information set.

[0032] In some embodiments, the executing entity can control at least one target chemical detection terminal identified through the detection terminal identification information set, and extract chemical status information from the multimodal target chemical information set corresponding to the at least one target chemical detection terminal to obtain a multimodal chemical status information set. The multimodal target chemical information in the multimodal target chemical information set can be local information related to the target chemical possessed by the target chemical detection terminal. The multimodal target chemical information in the multimodal target chemical information set may include, but is not limited to, at least one of the following: target chemical name, target chemical number, transportation equipment information related to the target chemical, involved enterprise information, involved personnel information, target chemical order information, target chemical video, image, concentration, and weight. The multimodal chemical status information in the multimodal chemical status information set can represent the status of the multimodal target chemical information. The multimodal chemical status information may include, but is not limited to, at least one of the following: target chemical concentration, weight, volume, transportation environment information, involved personnel information, and transportation vehicle information. The aforementioned multimodal state information of chemical substances can be obtained by preprocessing the aforementioned multimodal target chemical substance information.

[0033] Step 103: Perform data deduplication on the multimodal state information set of chemical products to obtain the deduplicated multimodal chemical product information set.

[0034] In some embodiments, the aforementioned executing entity may perform data deduplication processing on the aforementioned multimodal state information set of chemical articles to obtain a deduplicated multimodal chemical article information set.

[0035] Furthermore, in the process of adopting technical solutions to address the technical problems mentioned in the background, the following technical problems often arise: The multimodal state information set of chemical substances contains information about chemical substances with different modalities but the same meaning, resulting in a large amount of redundant data. This leads to low accuracy in the early warning and interception of subsequent tracked chemical substance information, wasting significant computational resources and prolonging the early warning and interception time. A conventional solution to this problem is to deduplicate the multimodal state information set of chemical substances by calculating the similarity between different modalities. However, this conventional solution still has the following problems: Determining redundant data in the multimodal state information set of chemical substances solely through similarity considers too few influencing factors, failing to accurately identify potential redundant information, resulting in low deduplication effectiveness, low accuracy in the identification and interception of subsequent tracked chemical substance information, wasting significant computational resources, and prolonging the identification and interception time. Considering the shortcomings of the conventional solution and combining the advantages / current state of multimodal data deduplication technology possessed by the applicant's research institute partners in this field, we have decided to adopt the following solution: In some optional implementations of certain embodiments, the above-mentioned data deduplication processing of the multimodal state information set of the chemical articles to obtain a deduplicated multimodal chemical article information set may include the following steps: The first step involves performing a cross-modal unified representation of the aforementioned multimodal state information set of chemical substances to obtain a multimodal represented chemical substance information set, and determining the spatial correlation set of the substance information among the multimodal represented chemical substance information included in the aforementioned multimodal represented chemical substance information set. The spatial correlation set of the substance information can characterize the spatial distribution characteristics and correlations among the aforementioned multimodal represented chemical substance information. In practice, the implementing entity can first perform a cross-modal unified representation of the aforementioned multimodal state information set of chemical substances to obtain the multimodal represented chemical substance information set. This multimodal represented chemical substance information set can be obtained by modally aligning the time-series data (represented using a Long Short-Term Memory neural network model), image data (represented using a convolutional neural network model), and text data (represented using a BERT (Bidirectional Encoder Representations from Transformers) model) of the different modalities of the chemical substance information set. Subsequently, the Euclidean distance between any two pieces of information representing multimodal chemical substances in the aforementioned multimodal chemical substance information set is determined, resulting in a set of Euclidean distance values, which serves as the multimodal spatial correlation set. Next, the standard deviation of the aforementioned multimodal chemical substance information set is determined as the data fluctuation coefficient. Then, the mean of the aforementioned multimodal spatial correlation set is determined as the spatial correlation mean. Next, the product of the exponential function value of e raised to the power of the data fluctuation coefficient and the spatial correlation mean is determined as the spatial model. Afterward, the aforementioned multimodal chemical substance information set is mapped to the aforementioned spatial model, resulting in a multimodal model chemical substance information set. Finally, using radial basis functions, a spatial correlation set of substance information for the multimodal model chemical substance information set is generated.

[0036] The second step is to perform weighted denoising on the above multimodal chemical product information set based on the spatial correlation set of the above product information to obtain the denoised chemical product information set.

[0037] As an example, the aforementioned execution entity can first select at least one multimodal chemical item information whose corresponding spatial correlation degree is greater than or equal to a preset spatial correlation degree threshold from the aforementioned multimodal chemical item information set, as the target multimodal information set. The preset spatial correlation degree threshold can be a pre-set critical value used to determine which denoising algorithm to use. Next, the target multimodal information set is weighted and denoised using the KNN (K-Nearest Neighbors) algorithm to obtain the denoised target multimodal information set. Then, the multimodal chemical item information set from which the target multimodal information set is removed is determined as the removed multimodal information set. Subsequently, the ratio of the spatial correlation degree of the item information corresponding to each removed multimodal item in the removed multimodal information set to the spatial correlation degree is determined as the spatial correlation degree, resulting in a set of spatial correlation degree values. The spatial correlation degree can be the sum of the spatial correlation degrees of the item information corresponding to the removed multimodal information set. Subsequently, using a weighted moving average filtering algorithm, the aforementioned multimodal information set is weighted and denoised according to the set of spatial removal weights, resulting in a denoised multimodal information set. Finally, the denoised multimodal information set and the denoised target multimodal information set are used to determine the denoised chemical information set.

[0038] The third step is to determine the autocorrelation distribution information set of the denoised chemical information set. The autocorrelation statistical distribution information in this set characterizes the degree of correlation between the denoised chemical information and itself at different time points. This determination can be performed using the Autocorrelation Function (ACF).

[0039] The fourth step involves dividing the denoised chemical product information set into blocks based on the aforementioned data autocorrelation distribution information set, resulting in a multimodal chemical product block information set. This multimodal chemical product block information set can be obtained by spatially and temporally dividing the denoised chemical product information set. In practice, the executing entity can first determine the partial autocorrelation value set of the time-series chemical product datasets included in the denoised chemical product information set using the partial autocorrelation function (PACF). Then, using an autoregressive moving average model, based on the aforementioned data autocorrelation distribution information set and the aforementioned partial autocorrelation value set, determine the set of autocorrelation orders of the time-series chemical product datasets. The autocorrelation order in this set characterizes the range of dependencies in the time-series chemical product data, i.e., the correlation between the time-series chemical product data and the preceding and following time-series chemical product datasets. Subsequently, the time-series chemical product dataset is segmented using the aforementioned autocorrelation order set of the products to obtain a non-time-series segmented information set, which serves as the multimodal chemical product segmented information set. Finally, the non-time-series chemical product data included in the denoised chemical product information set is segmented using the k-means clustering algorithm to obtain a non-time-series segmented information set, which also serves as the multimodal chemical product segmented information set.

[0040] The fifth step is to determine the multi-dimensional item feature information set of the aforementioned multimodal chemical item block information set. The multi-dimensional item feature information in this set can characterize the multi-dimensional features of the aforementioned multimodal chemical item block information. In practice, the executing entity can use principal component analysis to perform linear superposition processing on the aforementioned multimodal chemical item block information set to obtain the multi-dimensional item feature information set.

[0041] Step 6: Based on the aforementioned multi-dimensional item feature information set, perform multi-scale matching on the denoised chemical item information set to obtain a data source association feature information set. The data source association feature information in this set can characterize the association between the multi-modal chemical item block information and the multi-dimensional item feature information. This multi-scale matching can be performed using a cosine similarity-based matching algorithm.

[0042] Step 7: Based on the aforementioned data source associated feature information set, perform data clustering decomposition and reconstruction processing on the aforementioned denoised chemical product information set to obtain the reconstructed chemical product information set.

[0043] As an example, the aforementioned execution entity can first use a hierarchical clustering algorithm to perform data clustering on the denoised chemical product information set based on the associated feature information set of the aforementioned data source, obtaining a chemical product cluster set. Then, for each chemical product cluster in the aforementioned chemical product cluster set, the following decomposition and reconstruction steps are performed: First, using a fast wavelet decomposition algorithm, the denoised chemical product information groups included in the aforementioned chemical product cluster set are decomposed into decomposed chemical product information groups. Second, the decomposed chemical product information groups are reconstructed to obtain reconstructed chemical product information groups. Third, using an autoregressive redundant data prediction model, the reconstructed chemical product information group set is predicted step by step to obtain predicted deduplicated chemical product information groups, which serve as the reconstructed chemical product information set. The aforementioned autoregressive redundant data prediction model can be an autoregressive model that performs redundant data prediction on the input reconstructed chemical product information group. The aforementioned autoregressive redundant data prediction model can include: ARMA (Autoregressive Moving Average Model) and feature similarity redundancy prediction algorithms. An ARMA model can be used to predict time-series data included in a denoised chemical information set. A feature similarity redundancy prediction algorithm can be used to remove redundancy from non-time-series data with similarity greater than or equal to a preset similarity threshold included in the denoised chemical information set. The preset similarity threshold can be a pre-set threshold for determining whether to perform deduplication. For example, the preset similarity threshold could be 0.75.

[0044] Step 8 involves removing the reconstructed chemical information set from the aforementioned multimodal chemical information set to obtain a deduplicated multimodal chemical information set. Based on this deduplicated multimodal chemical information set, risk warning and interception are then performed on the aforementioned chemical information set to be tracked. It should be noted that the specific implementation of step 8 can be found in steps 104-106, and will not be elaborated upon here.

[0045] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem mentioned in the background art: "In the multimodal state information set of chemical substances, there are chemical substance information with different modalities but the same meaning, resulting in a large amount of redundant data. This leads to a low accuracy rate in the early warning and interception of subsequent chemical substance information to be tracked, wasting a large amount of computing resources and prolonging the early warning and interception time." The factors that lead to a low accuracy rate in the early warning and interception of subsequent chemical substance information to be tracked, wasting a large amount of computing resources, and prolonging the early warning and interception time are often as follows: The multimodal state information set of chemical substances contains chemical substance information with different modalities but the same meaning, resulting in a large amount of redundant data. This leads to a low accuracy rate in the early warning and interception of subsequent chemical substance information to be tracked, wasting a large amount of computing resources and prolonging the early warning and interception time. If these factors are solved, the accuracy rate of early warning and interception of information on chemical substances to be tracked can be improved, the waste of computing resources can be reduced, and the early warning and interception time can be shortened. To achieve this effect, this disclosure first performs cross-modal unified representation of the multimodal state information set of chemical products and determines the spatial correlation set of product information before denoising. The spatial correlation set of product information represents the spatial distribution characteristics and interrelationships of multimodal chemical product information. Different weighted denoising processes are applied to different product information spatial correlations, which can improve denoising efficiency and accuracy, as well as the quality of the denoised chemical product information set. Second, the denoised chemical product information set is segmented based on data autocorrelation distribution information and multi-dimensional product feature information is extracted. Segmentation can improve the accuracy of deduplication, and the extraction of multi-dimensional product feature information can extract potential feature information in the multimodal chemical product information set, so as to have a more comprehensive understanding of the multimodal chemical product information set. Subsequently, through the multi-dimensional product feature information set, multi-scale matching processing and data clustering decomposition and reconstruction processing are performed on the multimodal chemical product information set. Multi-scale matching processing can extract the inherent correlation between multimodal chemical product information, which is convenient for identifying potential redundant data among multimodal chemical product information. Then, the multimodal chemical information set is subjected to data clustering decomposition and reconstruction. Clustering can identify multimodal chemical information groups with high similarity into a cluster, which facilitates deduplication in each chemical cluster and improves the fineness of deduplication. The denoised chemical information groups included in each chemical cluster are decomposed and reconstructed, and the denoised chemical information groups are transformed into feature representations that are easier to analyze and process, which facilitates the extraction of key features of the data. Furthermore, the reconstructed chemical information set is predicted for redundant data through an autoregressive redundant data prediction model, which can improve the accuracy of redundant data prediction.Finally, by removing the reconstructed chemical information set from the aforementioned multimodal chemical information set, a deduplicated multimodal chemical information set is obtained. Based on this deduplicated multimodal chemical information set, risk warning and interception are performed on the aforementioned chemical information set to be tracked. This can improve deduplication efficiency and data quality of the deduplicated multimodal chemical information set, reduce the amount of data on chemical information with different modalities but the same meaning, improve the accuracy of warning and interception of chemical information to be tracked, reduce the waste of computing resources, and shorten the warning and interception time.

[0046] Step 104: Perform multimodal association recognition on the deduplicated multimodal chemical information set to obtain a multimodal association information set.

[0047] In some embodiments, the executing entity may perform multimodal association identification on the deduplicated multimodal chemical information set to obtain a multimodal association information set. The multimodal association information in the multimodal association information set may be information obtained by associating and connecting each deduplicated multimodal chemical information in the deduplicated multimodal chemical information set.

[0048] In some optional implementations of certain embodiments, the above-mentioned multimodal association identification of the deduplicated multimodal chemical information set to obtain a multimodal association information set may include the following steps: The first step involves performing text recognition on the audio data included in the deduplicated multimodal chemical information set to obtain an audio-text information set. The audio-text information in this set can be information describing the audio data in text form. The audio data can also be data describing the target chemical information in audio form.

[0049] The second step involves performing target recognition on the video and image data included in the deduplicated multimodal chemical information set to obtain a target detection information set. The target detection information in this set can be information about the target chemical items identified in the video or image data. For example, the target detection information could be the license plate number or vehicle type information of a transport vehicle. The video data can be data describing the target chemical item in video format. The image data can be data describing the target chemical item in image format. In practice, the target recognition can be performed using a convolutional neural network model.

[0050] The third step involves performing chemical entity recognition on the aforementioned audio-text information set, target detection information set, and deduplicated multimodal chemical information set to obtain a chemical entity set. The chemical entities in this set can be things or concepts related to the target chemical. For example, a chemical entity could be natural gas or transportation equipment used to transport the target chemical. In practice, the executing entity can first partition the data structure of the audio-text information set, target detection information set, and deduplicated multimodal chemical information set to obtain a structured chemical information set, a semi-structured chemical information set, and an unstructured chemical information set. Secondly, the field information set included in the structured chemical information set is identified as the structured chemical entity set. Then, regular expressions are used to extract entities from the semi-structured chemical information set to obtain a semi-structured chemical entity set. Finally, a named entity recognition model is used to identify entities in the unstructured chemical information set to obtain an unstructured chemical entity set. Finally, the unstructured chemical item entity set, the unstructured chemical item entity set, and the unstructured chemical item entity set are determined as the chemical item entity set.

[0051] The fourth step involves extracting entity relationship attributes from the aforementioned chemical entity set to obtain a static relationship information set, a dynamic relationship information set, and an attribute information set for chemical entities. The static relationship information in the static relationship information set can be information about the inherent relationships within the target chemical entity itself. The dynamic relationship information in the dynamic relationship information set can be information about relationships formed as the dosage of the target chemical changes. For example, the dynamic relationship information could be information about the dynamic relationship when the concentration of the target chemical reaches a preset concentration threshold, causing an explosion. The attribute information in the attribute information set can be information describing different aspects of the target chemical's performance. For example, the attribute information could include, but is not limited to, at least one of the following: reaction type, hazard type, storage conditions, and accident scenario information. In practice, the executing entity can use a relationship extraction model to extract entity relationship attributes from the aforementioned chemical entity set to obtain the static relationship information set, the dynamic relationship information set, and the attribute information set for chemical entities. The aforementioned relation extraction model can be the CasRel (Cascade and Residual Learning for Relational Triple Extraction) model.

[0052] The fifth step involves constructing a chemical knowledge graph from the aforementioned chemical entity set, static relationship information set, dynamic relationship information set, and attribute information set. This chemical knowledge graph can be a graph structure that displays information about the target chemical in graph form. The construction of the chemical knowledge graph can be achieved by storing the aforementioned chemical entity set, static relationship information set, dynamic relationship information set, and attribute information set as triples in a graph database. These triples can be <chemical head entity, relationship information, chemical tail entity> and <chemical entity, attribute, chemical entity attribute information>. The chemical head entity and chemical tail entity can be entities within the aforementioned chemical entity set. The relationship information can be the relationship information within the aforementioned static relationship information set and dynamic relationship information set.

[0053] Step 6: Perform entity relation disambiguation processing on the aforementioned chemical product knowledge graph to obtain a disambiguated chemical product knowledge graph. This disambiguated chemical product knowledge graph can be obtained by replacing multiple entities and relationships in the aforementioned chemical product knowledge graph with the same entity and relationship. In practice, the executing entity can perform feature encoding on the entity set and relationship set included in the aforementioned chemical product knowledge graph to obtain an entity feature vector set and a relationship feature vector set. Then, determine the feature similarity between any two entity feature vectors in the entity feature vector set to obtain an entity similarity set, and determine the feature similarity between any two relationship feature vectors in the relationship feature vector set to obtain a relationship similarity set. Finally, perform entity relation disambiguation on at least one entity and at least one relationship corresponding to at least one entity similarity and at least one relationship similarity greater than or equal to a preset similarity threshold to obtain the disambiguated chemical product knowledge graph.

[0054] The seventh step involves performing association mining on the disambiguated chemical product knowledge graph to obtain a multimodal association information set. This association mining can be performed using a path ranking algorithm.

[0055] In some optional implementations of certain embodiments, the above-mentioned correlation mining of the disambiguated chemical knowledge graph to obtain a multimodal correlation information set may include the following steps: The first step involves sub-partitioning the disambiguated chemical knowledge graph to obtain a target chemical sub-graph set. The target chemical sub-graphs within this set can be obtained by community partitioning of the disambiguated chemical knowledge graph. In practice, the executing entity can utilize the Louvain algorithm to perform sub-partitioning on the disambiguated chemical graph to obtain the target chemical sub-graph set.

[0056] The second step involves identifying key nodes in each target chemical sub-graph within the aforementioned target chemical sub-graph set to generate target key node groups, thus obtaining a target key node set. The target key nodes in this set represent nodes that play a crucial role in the target chemical sub-graph, i.e., central nodes. In practice, the executing entity can utilize the PageRank algorithm to identify key nodes in each target chemical sub-graph within the aforementioned target chemical sub-graph set to generate target key node groups, thus obtaining a target key node set.

[0057] The third step involves performing path association mining on the disambiguated chemical product knowledge graph based on the aforementioned set of key target nodes, thereby obtaining a set of indirect entity association information. This set of indirect entity association information can be information on potential causal chains within the disambiguated chemical product knowledge graph. For example, the executing entity can utilize a path analysis algorithm to perform path association mining on the disambiguated chemical product knowledge graph based on the aforementioned set of key target nodes, thereby obtaining the set of indirect entity association information. This path analysis algorithm can be one or a combination of algorithms such as breadth-first search, depth-first search, Dijkstra's algorithm, and path ranking algorithms.

[0058] The fourth step involves predicting link relationships in the disambiguated chemical knowledge graph to obtain a predicted relationship information set. This predicted relationship information set can include previously undiscovered relationships predicted through a graph neural network model. In practice, the executing entity can input the disambiguated chemical knowledge graph into a graph neural network model for entity embedding learning to obtain the predicted relationship information set.

[0059] The fifth step involves using an association rule engine to mine association rules within the disambiguated chemical knowledge graph, yielding a set of inferential association information. This inferential association information can be information on reaction and temporal relationships between potential chemicals discovered through the association rule engine. The association rule engine can be a pre-defined rule engine built upon existing knowledge of hazardous chemicals. For example, an inferential association could be: "Target chemical information A has oxidizing properties, and target chemical information B has reducing properties; therefore, it can be inferred that A and B may cause an explosion, indicating a correlation between A and B that could trigger an explosion."

[0060] The sixth step is to perform deduplication on the above-mentioned indirect relationship information set, predicted relationship information set, and inferred relationship information set to obtain a deduplicated relationship information set, which serves as the multimodal relationship information set.

[0061] It should be noted that by using the rule-based reasoning-based association mining in step five, known associations that conform to the rules can be extracted with high accuracy and interpretability; the association mining based on graph algorithm reasoning in step four can capture structurally dependent associations; and the association mining based on machine learning in steps one through three can mine unknown associations and can automatically adapt to the data. The three association mining methods complement each other and can improve the accuracy and comprehensiveness of association mining.

[0062] Step 105: Based on the multimodal association information set, perform chemical risk identification on the deduplicated multimodal chemical information set to obtain the chemical information set to be tracked.

[0063] In some embodiments, the aforementioned executing entity may perform chemical risk identification on the deduplicated multimodal chemical information set based on the aforementioned multimodal association information set to obtain a chemical information set to be tracked. The chemical information to be tracked in the aforementioned chemical information set may be information on chemicals identified as hazardous and requiring tracking and interception.

[0064] In some optional implementations of certain embodiments, the process of identifying chemical risks in the deduplicated multimodal chemical information set based on the multimodal association information set to obtain the chemical information set to be tracked may include the following steps: The first step involves using a target chemical detection model to detect target chemicals in the video and image data included in the deduplicated multimodal chemical information set, thereby obtaining the target chemical information set. The target chemical information set can include information related to the target chemical. This target chemical information set may include: the name, category, and transport vehicle information of the target chemical. The target chemical detection model can be a deep neural network model that detects the input video and image data and outputs detection results. For example, the target chemical detection model could be a Faster R-CNN (Faster Region-based Convolutional Neural Networks) model.

[0065] The second step is to determine the pedestrian behavior information set corresponding to the aforementioned target chemical information set. This pedestrian behavior information set can include information on the behavior of pedestrians surrounding the target chemical information set. This determination can be performed using a pedestrian behavior posture recognition model. This model can be a model that performs pedestrian posture recognition on the pedestrian videos included in the input target chemical information set. For example, the pedestrian behavior posture recognition model could be the DeepPose model.

[0066] The third step is to extract pedestrian motion optical flow information from the aforementioned pedestrian behavior information set to obtain a pedestrian motion optical flow information set. This set can be the pedestrian movement information within the time frame corresponding to the video data being captured.

[0067] As an example, the aforementioned executing entity can use the Lucas-Kanade optical flow method to extract pedestrian motion optical flow information from the aforementioned pedestrian behavior information set, thereby obtaining the pedestrian motion optical flow information set.

[0068] The fourth step involves extracting spatiotemporal information from the aforementioned pedestrian behavior information set to obtain a pedestrian spatiotemporal information set. This set represents the temporal and spatial aspects of pedestrian behavior. In practice, the implementing entity can first input the aforementioned pedestrian behavior information set into EfficientNet. The B1 model is used to obtain a behavioral feature map set. This behavioral feature map set is then input into a channel-space-temporal multi-attention mechanism network to obtain a pedestrian spatiotemporal information set. The channel-space-temporal multi-attention mechanism network can extract spatiotemporal information through the following steps: First, the behavioral feature map set is input into an average pooling layer and a max pooling layer respectively for feature fusion, and then concatenated with the behavioral feature map set to obtain a channel attention feature map. Second, the channel attention feature map is expanded along the feature channel direction and then input into an average pooling layer and a max pooling layer respectively for frame-level feature fusion to obtain a spatial aggregation feature map. Third, the spatial aggregation feature map is input into a standard convolutional layer for channel scaling and then multiplied with the channel attention feature map to obtain a spatial attention feature map. Finally, the spatial attention feature map undergoes three sets of convolutional processing for scaling transformation to obtain a first transform attention feature map, a second transform attention feature map, and a third transform attention feature map. Then, the first and second transform attention feature maps are multiplied pixel-by-pixel, followed by a pixel-by-pixel multiplication with the third transform attention feature map, and then fed into a softmax activation function to obtain a temporal attention feature map. Next, the temporal attention feature map is fed into a convolutional layer to reconstruct a temporal transform attention feature map with the same size as the behavioral feature map. Finally, the temporal transform attention feature map and the spatial attention feature map are fused and fed into a fully connected layer to obtain the pedestrian spatiotemporal information set.

[0069] The fifth step involves identifying abnormal pedestrians using the aforementioned pedestrian motion optical flow information set and pedestrian spatiotemporal information set to obtain an abnormal behavior information set. This abnormal behavior information set characterizes pedestrians exhibiting unusual behavior. For example, it could indicate multiple people rapidly moving around a vehicle in a non-transportation equipment parking area, pedestrians frequently turning their heads to observe surveillance cameras, or pedestrians briefly gathering and then dispersing. This brief gathering could be less than 5 minutes. In practice, the executing entity can first input the aforementioned pedestrian motion optical flow information set and pedestrian spatiotemporal information set into an average pooling layer, then into a Dropout layer, and finally into a linear layer to obtain motion optical flow feature maps and spatiotemporal feature maps. Then, the motion optical flow feature maps and the spatiotemporal feature maps are input into a TFT (Temporal Fusion Transformer) model to obtain the abnormal behavior information set.

[0070] Step 6: Based on the aforementioned abnormal behavior information set, determine the target chemical risk level information set for the aforementioned target chemical information set. The target chemical risk level information in the aforementioned target chemical risk level information set characterizes the risk level of the aforementioned target chemical information. In practice, the executing entity can determine the set of abnormal behavior probability values ​​corresponding to the aforementioned abnormal behavior information set as the target chemical risk level information set for the aforementioned target chemical information set.

[0071] Step 7: Based on the aforementioned multimodal association information set and the aforementioned target chemical risk level information set, perform chemical risk identification on the aforementioned target chemical information set to obtain the chemical information set to be tracked. The chemical information to be tracked in the aforementioned chemical information set may include information on hazardous chemicals that do not comply with national regulations or are transported privately.

[0072] As an example, the aforementioned implementing entity may first perform an association search on the aforementioned multimodal association information set based on the aforementioned target chemical information set to obtain transportation information related to the target chemical information set, which will serve as the chemical association information set. The aforementioned chemical association information set may include, but is not limited to, at least one of the following: transportation route, chemical name and category information, and information on companies involved in the transportation. Finally, from the aforementioned target chemical risk level information set, at least one target chemical information corresponding to at least one target chemical risk level information greater than or equal to a preset risk level threshold is selected to obtain the chemical information set to be tracked. The aforementioned preset risk level threshold may be a pre-set threshold value used to determine whether a chemical is included in the chemical information set to be tracked. Therefore, by detecting pedestrian behavior information related to the target chemical information set, such as pedestrian motion optical flow information and pedestrian spatiotemporal information, it is possible to determine whether there are any anomalies in the pedestrian behavior information and the degree of anomaly risk. This allows for the analysis and determination of the pedestrian's behavioral intentions, making pedestrian anomaly detection an effective auxiliary information for determining the risk level of the target chemical. Furthermore, by using a multimodal association information set, it is possible to further determine whether the target chemical information set is indeed the chemical information to be tracked. This can further improve the accuracy and comprehensiveness of identifying the chemical information to be tracked, enabling precise interception of the chemical information to be tracked, reducing the harm caused by the target chemical information, and improving social security.

[0073] Step 106: Perform trajectory tracking and early warning on the set of chemical substances to be tracked to obtain a set of chemical substance trajectory early warning information. Based on the set of chemical substance trajectory early warning information, control the set of chemical substance early warning devices and the set of chemical substance interception devices to perform early warning and interception on the set of chemical substances to be tracked corresponding to the set of chemical substances to be tracked.

[0074] In some embodiments, the aforementioned executing entity can perform trajectory tracking and early warning on the aforementioned set of chemical substances to be tracked, obtaining a chemical substance trajectory early warning information set, and, based on the aforementioned chemical substance trajectory early warning information set, control a set of chemical substance early warning devices and a set of chemical substance interception devices to perform early warning and interception on the set of chemical substances to be tracked corresponding to the aforementioned set of chemical substance information. The chemical substance trajectory early warning information in the aforementioned chemical substance trajectory early warning information set can be information providing early warning of the transportation trajectory of the chemical substances to be tracked corresponding to the aforementioned chemical substance information. The chemical substance interception devices in the aforementioned set of chemical substance interception devices can be hardware devices used to intercept the chemical substances to be tracked. The chemical substance early warning devices in the aforementioned set of chemical substance early warning devices can be devices that provide early warning of the chemical substances to be tracked through sound and color. For example, the aforementioned transportation interception device can be a roadblock machine.

[0075] In some optional implementations of certain embodiments, the above-mentioned trajectory tracking and early warning of the above-mentioned set of information on the chemicals to be tracked, to obtain a chemical trajectory early warning information set, may include the following steps: The first step is to determine the environmental information set of the transported chemicals for the aforementioned set of information to be tracked. This environmental information set can include details about the transportation environment in which the chemical information set is transported. This environmental information may include: temperature and humidity during vehicle transport and storage, concentration information of the chemical information to be tracked, and transportation route information. In practice, the implementing entity can obtain this environmental information set from sensors on the transport equipment carrying the chemical information set and from road cameras.

[0076] The second step involves performing transport path prediction processing on the aforementioned set of chemical items to be tracked, based on the historical trajectory information and the aforementioned transport environment information, to obtain a predicted trajectory transport path information set. The predicted trajectory transport path information in this set can be transport path information obtained by predicting the transport path of the aforementioned set of chemical items to be tracked. The historical trajectory information can be the transport path information of the transport equipment carrying the set of chemical items to be tracked, located prior to the current time. As an example, the executing entity can utilize a spatiotemporal graph neural network to perform transport path prediction processing on the aforementioned set of chemical items to be tracked, based on the historical trajectory information and the aforementioned transport environment information, to obtain the predicted trajectory transport path information set.

[0077] The third step involves determining, based on the aforementioned multimodal association information set, the target user trajectory information set and the target user communication information set associated with the aforementioned chemical information set to be tracked. Specifically, the target user trajectory information in the target user trajectory information set can be information about the behavioral trajectories of users associated with the aforementioned chemical information to be tracked. These associated users can include, but are not limited to, at least one of the following: users transporting the chemical information to be tracked, users receiving the chemical information to be tracked, and users producing the chemical information to be tracked. The target user communication information in the target user communication information set can be information about communication connections between the associated users and other users.

[0078] As an example, the aforementioned executing entity may first search the multimodal association information set to obtain target user information associated with the information set of the chemical substances to be tracked. Then, it may perform trajectory tracking on the target user information to obtain a target user trajectory information set. Finally, based on the target user information, it may perform a communication connection search on the multimodal association information set to obtain a target user communication information set.

[0079] The fourth step involves processing the predicted trajectory transport path information set based on the target user trajectory information set and the target user communication information set to obtain a chemical product trajectory warning information set.

[0080] As an example, the aforementioned execution entity can first perform multimodal fusion of the target user trajectory information set and the target user communication information set through timestamp alignment to obtain fused information. Then, using the A* search algorithm, the predicted trajectory transport path information is dynamically adjusted to obtain an adjusted transport path information set. Next, for each adjusted transport path information in the adjusted transport path information set, in response to determining the preset warning areas included in the preset warning area set that the adjusted transport path information passes through, a path warning information set is generated, and the number of preset warning areas passed through by the adjusted transport path is determined to generate a set of number of warning areas passed through, resulting in a set of number of warning areas passed through. The preset warning areas in the preset warning area set can be pre-defined areas that require key monitoring and warning. For example, the preset warning areas can be densely populated areas or military bases. Finally, based on the set of number of warning areas, the path warning information set is classified into warning levels to obtain a level-classified warning information set, which serves as the chemical trajectory warning information set.

[0081] Furthermore, in the process of adopting technical solutions to address the technical problems mentioned in the background, the following technical problems often arise: During the identification of target chemical information, the risk level of the target chemical information cannot be accurately assessed, resulting in a low interception accuracy rate. This leads to the inability of chemical warning and interception equipment sets to provide timely warnings and interception of target chemical information, increasing the probability of accidents involving the target chemical information and raising the damage rate of transportation interception devices. Conventional solutions to these technical problems generally involve using quantitative methods to predict the risk information of the chemical information to be tracked based on historical data, and then intercepting it. However, these conventional solutions still have the following problems: prediction based solely on historical data does not consider the differences between the chemical information to be tracked and historical data, resulting in poor accuracy of the predicted risk information. This leads to low warning and interception accuracy rates for chemical warning and interception equipment sets, increasing the probability of accidents involving the target chemical and reducing public safety. Considering the shortcomings of the conventional solutions and combining the technological advantages / current status of the applicant's research institute partners in this field regarding the warning and interception of tracked chemicals, we have decided to adopt the following solution: In some optional implementations of certain embodiments, the above-mentioned control of the chemical warning device set and the chemical interception device set based on the chemical trajectory warning information set to perform warning interception on the set of chemical substances to be tracked corresponding to the chemical substance information set may include the following steps: The first step is to determine the chemical type information and chemical operation status information for each piece of information in the aforementioned set of information to be tracked chemicals, thereby obtaining a chemical type information set and a chemical operation status information set. The chemical type information can be the information on the chemical category to which the aforementioned chemical item belongs. The chemical operation status information can be the information on the storage environment in which the aforementioned set of information to be tracked chemicals is stored. The chemical operation status information may include, but is not limited to, at least one of the following: storage temperature and humidity information, packaging information of the chemical item to be tracked, and transport volume.

[0082] The second step involves performing driving status identification processing on the set of transport drivers corresponding to the aforementioned set of information on the traceable chemicals, resulting in a set of driving status information. This driving status information set characterizes the state of the transport driver's operation of the transport equipment. In practice, the executing entity can use the YOLOv7 (You Only LookOnce) model to identify the state of the transport drivers corresponding to the aforementioned set of information on the traceable chemicals, thereby obtaining the driving status information set.

[0083] The third step is to determine the vehicle transportation status information set corresponding to the aforementioned set of information on the chemicals to be tracked. This vehicle transportation status information set may include road traffic information about the transport vehicles traveling on the road. This vehicle transportation status information may include, but is not limited to, at least one of the following: road information, traffic condition information, distance information from adjacent vehicles, the proportion of vehicles violating traffic rules, and weather information.

[0084] The fourth step is to determine the risk weight value set for the aforementioned chemical product type information set, chemical product operational status information set, driving status information set, and vehicle transportation status information set. The risk weight values ​​in this risk weight value set can represent the proportion of any information in each of these sets. In practice, the implementing entity can first determine the subjective weight judgment matrix for the aforementioned chemical product type information set, chemical product operational status information set, driving status information set, and vehicle transportation status information set using the five-scale method in the Analytic Hierarchy Process (AHP). The five-scale method in the AHP can be a method of determining weight values ​​using five scales. The subjective weight judgment matrix can be a weight matrix composed of the weights of the aforementioned chemical product type information set, chemical product operational status information set, driving status information set, and vehicle transportation status information set determined by multiple experts using the five-scale method in the AHP. Next, the largest eigenvalue and its corresponding eigenvector of the subjective weight judgment matrix are calculated, and the elements of the eigenvector are normalized to obtain the first initial weight value set. Then, using the entropy weight method, the second initial weight value set is determined for the chemical product type information set, the chemical product operation status information set, the driving status information set, and the vehicle transportation status information set. Afterward, an objective function is constructed that minimizes the sum of squared deviations of the first initial weight value set, the second initial weight value set, the first assigned weight parameter, and the second assigned weight parameter. Subsequently, the objective function for minimizing the sum of squared deviations is solved to obtain the specific numerical sets corresponding to the first and second assigned weight parameters, which serve as the first and second assigned weight values. Finally, the product of the first initial weight value set and the first assigned weight value, and the product of the second initial weight value set and the second assigned weight value, are summed to obtain the risk weight value set.

[0085] The fifth step involves inputting the aforementioned chemical type information set, chemical operation status information set, driving status information set, vehicle transportation status information set, and risk weight value set into a multi-indicator target chemical risk warning and prediction model to obtain a target chemical warning information set. The target chemical warning information in this set represents the risk level of the tracked chemical information and is used to generate early warnings. The multi-indicator target chemical risk warning and prediction model can be a function that assesses the probability of potential risks caused by the tracked chemical information set using the first and second initial weight value sets corresponding to the chemical type information set, chemical operation status information set, driving status information set, and vehicle transportation status information set. In practice, the implementing entity can first determine a risk prediction and assessment matrix based on expert experience for the aforementioned chemical type information set, chemical operation status information set, driving status information set, and vehicle transportation status information set. The risk prediction and assessment matrix has the following structure: the number of rows can be the number corresponding to the chemical product type information set, the chemical product operation status information set, the driving status information set, and the vehicle transportation status information set; the number of columns can be the number of experts. Next, the minimum value corresponding to the chemical product type information set, the chemical product operation status information set, the driving status information set, and the vehicle transportation status information set is determined as the target value set. Then, the risk prediction and assessment matrix raised to the power of 0.9 is determined as the utility value matrix. Subsequently, the difference between 1 and the value of an exponential function with base e and the negative of the product of the difference between the target value set and the risk prediction and assessment matrix and 0.0133 is determined as the regret-joy function value matrix. This regret-joy function value matrix characterizes the degree of regret or joy generated when choosing different options. Finally, the sum of the utility value matrix and the regret-joy function value matrix is ​​determined as the perceived utility value matrix. Subsequently, the ratio of the sum of the products of the perceived utility value matrix and the aforementioned risk weight value set to the number of columns is determined to establish the transportation warning value set. Finally, the sum of the products of the aforementioned transportation warning value set and the risk weight value set is determined as the target chemical warning information set.

[0086] Step 6: Classify the aforementioned target chemical warning information set and the aforementioned chemical trajectory warning information set into warning levels to obtain a warning level information set. The warning level information in this set represents the warning level for the chemical being tracked. The risk level information can be obtained by weighted summing of each target chemical warning information in the target chemical warning information set and the corresponding chemical trajectory warning information in the chemical trajectory warning information set. The warning level information set may include: red (highest warning level), orange (relatively high warning level), yellow (medium warning level), blue (relatively low warning level), and green (safe warning level).

[0087] Step 7: Based on the aforementioned warning level information set, control the aforementioned chemical warning equipment set and the aforementioned chemical interception equipment set to perform different levels of interception processing on the aforementioned set of chemicals to be tracked. In practice, the implementing entity can first determine at least one chemical warning device and at least one chemical interception device corresponding to the aforementioned risk level information set from the aforementioned chemical warning equipment set and the aforementioned chemical interception equipment set, as the level warning equipment set and the level interception equipment set. Then, control the aforementioned level warning equipment set and the aforementioned level interception equipment set to perform different levels of interception processing on the aforementioned set of chemicals to be tracked.

[0088] The above-described technical solution and its related content, as an inventive point of this disclosure, solve the technical problem mentioned in the background art: "In the process of identifying target chemical information, the risk level of the target chemical information cannot be accurately assessed, resulting in a low interception accuracy rate of the target chemical information. This leads to the inability of the chemical warning device set and the interception device set to promptly warn and intercept the target chemical, increasing the probability of accidents caused by the target chemical information and increasing the damage rate of the transportation interception device." The factors that cause the chemical warning device set and the interception device set to fail to promptly warn and intercept the target chemical, increasing the probability of accidents caused by the target chemical information and increasing the damage rate of the transportation interception device are often as follows: In the process of identifying target chemical information, the risk level of the target chemical information cannot be accurately assessed, resulting in a low interception accuracy rate of the target chemical information. This leads to the inability of the chemical warning device set and the interception device set to promptly warn and intercept the target chemical, increasing the probability of accidents caused by the target chemical information and increasing the damage rate of the transportation interception device. If the above factors are solved, the accuracy of timely warning and interception of target chemical by the chemical warning device set and the interception device set can be improved, the probability of accidents caused by the target chemical information reduced, and the damage rate of the transportation interception device lowered. To achieve this effect, this disclosure first determines various current and historical information about the chemical substance to be tracked, considering both historical and current data. This allows for accurate determination of the current environmental information of the chemical substance, facilitating subsequent precise assessment of its risk level. Second, a combination of subjective and objective assignment methods is used to determine the risk weight values ​​for the chemical substance to be tracked. This avoids, to some extent, the discrepancies in weight values ​​caused by a single assignment method, improving the accuracy of the risk weight values ​​and making them more consistent with the current information of the chemical substance to be tracked. Third, a multi-indicator target chemical substance risk warning and prediction model is used to determine the target chemical substance warning information set. This model can, to some extent, avoid the subjectivity and variability caused by subjective assessment, thus improving the accuracy of the target chemical substance warning information set. Finally, the warning information set and the chemical substance trajectory warning information set are classified into warning levels. By classifying the warning information levels based on the risk and trajectory risk of the target chemical substance, the accuracy of the warning level classification can be improved. Finally, by controlling the aforementioned chemical warning equipment set and chemical interception equipment set through the warning level information set, the aforementioned set of chemicals to be tracked can be intercepted at different levels. By dividing different warning levels, chemical warning equipment and chemical interception equipment can be rationally arranged for accurate warning and interception, which can improve the rationality of equipment allocation, reduce the damage rate of chemical warning equipment and chemical interception equipment, improve the accuracy of warning and interception, improve social safety, and reduce the probability of accidents involving target chemicals.

[0089] Further reference Figure 2 As an implementation of the methods shown in the above figures, this disclosure provides some embodiments of a chemical substance early warning and interception device, which are similar to... Figure 1 Corresponding to the method embodiments shown, this chemical warning and interception device can be specifically applied to various electronic devices.

[0090] like Figure 2 As shown, a chemical substance early warning and interception device 200 includes: an identity recognition unit 201, a chemical substance status information extraction unit 202, a data deduplication unit 203, a multimodal association recognition unit 204, a chemical substance risk recognition unit 205, and an early warning and interception unit 206. The identity recognition unit 201 is configured to: in response to a detected data connection request information set of a target chemical substance detection terminal set, perform identity recognition on the target chemical substance detection terminal set to obtain a detection terminal identity recognition information set. The chemical substance status information extraction unit 202 is configured to: control at least one target chemical substance detection terminal represented by the detection terminal identity recognition information set to extract chemical substance status information from the multimodal target chemical substance information set corresponding to the at least one target chemical substance detection terminal to obtain a chemical substance multimodal status information set. The data deduplication unit 203 is configured to: perform data deduplication processing on the chemical substance multimodal status information set to obtain a deduplicated multimodal chemical substance information set. The multimodal association identification unit 204 is configured to perform multimodal association identification on the deduplicated multimodal chemical information set to obtain a multimodal association information set. The chemical risk identification unit 205 is configured to perform chemical risk identification on the deduplicated multimodal chemical information set based on the multimodal association information set to obtain a chemical information set to be tracked. The early warning and interception unit 206 is configured to perform trajectory tracking and early warning on the chemical information set to be tracked to obtain a chemical trajectory early warning information set, and control the chemical early warning device set and the chemical interception device set based on the chemical trajectory early warning information set to perform early warning and interception on the chemical information set to be tracked corresponding to the chemical information set to be tracked.

[0091] It is understandable that the units described in the chemical warning and interception device 200 are consistent with the reference... Figure 1 The steps in the described method correspond to each other. Therefore, the operations, features, and beneficial effects described above for the method also apply to the chemical warning and interception device 200 and the units contained therein, and will not be repeated here.

[0092] The following is for reference. Figure 3 It shows a schematic diagram of the structure of an electronic device (e.g., an electronic device) 300 suitable for implementing some embodiments of the present disclosure.Figure 3 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments of this disclosure.

[0093] like Figure 3 As shown, the electronic device 300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 302 or a program loaded from a storage device 308 into a random access memory (RAM) 303. The RAM 303 also stores various programs and data required for the operation of the electronic device 300. The processing unit 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to the bus 304.

[0094] Typically, the following devices can be connected to I / O interface 305: input devices 306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 309. Communication device 309 allows electronic device 300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 3 An electronic device 300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively. Figure 3 Each box shown can represent a device or multiple devices as needed.

[0095] In particular, according to some embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, some embodiments of this disclosure include a computer program product comprising a computer program carried on a computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 309, or installed from storage device 308, or installed from ROM 302. When the computer program is executed by processing device 301, it performs the functions defined in the methods of some embodiments of this disclosure.

[0096] It should be noted that, in some embodiments of this disclosure, the computer-readable medium described above may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In some embodiments of this disclosure, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In some embodiments of this disclosure, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals may take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0097] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0098] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device. The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: respond to detecting a data connection request information set of a target chemical substance detection terminal set, identify the target chemical substance detection terminal set, and obtain a detection terminal identification information set; control at least one target chemical substance detection terminal represented by the detection terminal identification information set to pass identification, extract chemical substance state information from the multimodal target chemical substance information set corresponding to the at least one target chemical substance detection terminal, and obtain a chemical substance multimodal state information set; and process the chemical substance multimodal state information... The information set is deduplicated to obtain a deduplicated multimodal chemical information set; multimodal association identification is performed on the deduplicated multimodal chemical information set to obtain a multimodal association information set; based on the multimodal association information set, chemical risk identification is performed on the deduplicated multimodal chemical information set to obtain a chemical information set to be tracked; trajectory tracking and early warning are performed on the chemical information set to be tracked to obtain a chemical trajectory early warning information set; and based on the chemical trajectory early warning information set, a chemical early warning device set and a chemical interception device set are controlled to perform early warning and interception on the chemical information set to be tracked.

[0099] Computer program code for performing operations of some embodiments of this disclosure can be written in one or more programming languages ​​or a combination thereof, including object-oriented programming languages ​​such as Java, Smalltalk, and C++, and conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0100] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0101] The units described in some embodiments of this disclosure can be implemented in software or hardware. The described units can also be housed in a processor; for example, a processor may be described as including an identity recognition unit, a chemical substance status information extraction unit, a data deduplication unit, a multimodal association recognition unit, a chemical substance risk recognition unit, and a warning and interception unit. The names of these units do not necessarily limit the specific unit; for example, the identity recognition unit may also be described as "a unit that, in response to detecting a data connection request information set of a target chemical substance detection terminal set, performs identity recognition on the target chemical substance detection terminal set to obtain a detection terminal identity recognition information set."

[0102] The functions described above in this document can be performed at least in part by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: field-programmable gate arrays (FPGAs), application-specific integrated circuits (ASICs), application-specific standard products (ASSPs), system-on-a-chip (SoCs), complex programmable logic devices (CPLDs), and so on.

[0103] The above description is merely a selection of preferred embodiments of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in the embodiments of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in the embodiments of this disclosure.

Claims

1. A method for early warning and interception of chemical substances, comprising: In response to the detection of a set of data connection request information for a set of target chemical detection terminals, the target chemical detection terminal set is identified to obtain a set of detection terminal identification information. The system controls the identification information set of the detection terminal to represent at least one target chemical detection terminal that has passed the identification, and extracts chemical status information from the multimodal target chemical information set corresponding to the at least one target chemical detection terminal to obtain a multimodal chemical status information set; The multimodal state information set of the chemical articles is deduplicated to obtain a deduplicated multimodal chemical article information set. Multimodal association recognition is performed on the deduplicated multimodal chemical information set to obtain a multimodal association information set; Based on the multimodal association information set, chemical risk identification is performed on the deduplicated multimodal chemical information set to obtain the chemical information set to be tracked; The system performs trajectory tracking and early warning on the set of chemical substances to be tracked, thereby obtaining a set of chemical substance trajectory early warning information. Based on the set of chemical substance trajectory early warning information, the system controls a set of chemical substance early warning devices and a set of chemical substance interception devices to perform early warning and interception on the set of chemical substances to be tracked corresponding to the set of chemical substances to be tracked.

2. The method according to claim 1, wherein, The step of responding to the detection of a data connection request information set for a target chemical detection terminal set, and performing identity verification on the target chemical detection terminal set to obtain a detection terminal identity verification information set, includes: The identity information is extracted from the data connection request information set to obtain the terminal digital certificate information set, the terminal current dynamic token set, and the terminal request signature information set; The terminal digital certificate information set is verified to obtain a digital certificate verification result information set; From the set of target chemical detection terminals, at least one target chemical detection terminal whose digital certificate verification result information indicates that it has passed the verification is selected, thus obtaining the selected set of target chemical detection terminals; Based on the dynamic token key set and the current timestamp set included in the terminal digital certificate information set corresponding to the selected target chemical substance detection terminal set, a target dynamic token set to be verified is generated. The target set of dynamic tokens to be verified and the current set of dynamic tokens of the terminal are compared to obtain a set of dynamic token comparison results. The terminal request signature information set is subjected to data anti-tampering verification to obtain a data anti-tampering verification information set; Based on the dynamic token comparison result information set and the data anti-tampering verification information set, a detection terminal identity recognition information set is generated.

3. The method according to claim 1, wherein, The process of performing trajectory tracking and early warning on the set of information about the chemicals to be tracked, resulting in a chemical trajectory early warning information set, includes: Determine the information set of the transport environment of the chemical substances to be tracked; Based on the historical trajectory information of the chemical information set to be tracked and the transportation environment information set of the chemical, the transportation path prediction process is performed on the chemical information set to be tracked to obtain the predicted trajectory transportation path information set. Based on the multimodal association information set, determine the target user trajectory information set and the target user communication information set associated with the chemical information set to be tracked; Based on the target user trajectory information set and the target user communication information set, trajectory warning processing is performed on the predicted trajectory transport path information set to obtain a chemical product trajectory warning information set.

4. The method according to claim 1, wherein, The step of performing chemical risk identification on the deduplicated multimodal chemical information set based on the multimodal association information set to obtain the chemical information set to be tracked includes: Using a target chemical detection model, target chemical detection is performed on the video data and image data included in the deduplicated multimodal chemical information set to obtain the target chemical information set. Determine the pedestrian behavior information set corresponding to the target chemical information set; Pedestrian motion optical flow information is extracted from the pedestrian behavior information set to obtain the pedestrian motion optical flow information set; Spatiotemporal information is extracted from the pedestrian behavior information set to obtain the pedestrian spatiotemporal information set; Anomaly identification is performed on the pedestrian motion optical flow information set and the pedestrian spatiotemporal information set to obtain an abnormal behavior information set; Based on the abnormal behavior information set, determine the target chemical risk level information set of the target chemical information set; Based on the multimodal association information set and the target chemical risk level information set, chemical risk identification is performed on the target chemical information set to obtain the chemical information set to be tracked.

5. The method according to claim 1, wherein, The process of performing multimodal association recognition on the deduplicated multimodal chemical information set to obtain a multimodal association information set includes: Text recognition is performed on the audio data included in the deduplicated multimodal chemical information set to obtain an audio-text information set; Target recognition is performed on the video data and image data included in the deduplicated multimodal chemical information set to obtain a target detection information set; Chemical entity recognition is performed on the audio text information set, the target detection information set, and the deduplicated multimodal chemical information set to obtain a chemical entity set; Entity relationship attributes are extracted from the chemical entity set to obtain a static relationship information set, a dynamic relationship information set, and an attribute information set of chemical entities. A chemical knowledge graph is constructed from the chemical entity set, the static relationship information set of chemical entities, the dynamic relationship information set of chemical entities, and the attribute information set of chemical entities to obtain a chemical knowledge graph. The chemical product knowledge graph is subjected to entity relation disambiguation processing to obtain a disambiguated chemical product knowledge graph. The disambiguated chemical knowledge graph is subjected to association mining to obtain a multimodal association information set.

6. The method according to claim 5, wherein, The process of mining association relationships in the disambiguated chemical knowledge graph yields a multimodal association information set, including: The disambiguated chemical knowledge graph is divided into subgraphs to obtain the target chemical subgraph set. For each target chemical sub-map in the target chemical sub-map set, key nodes are identified to generate target key node groups, thus obtaining a target key node group set; Based on the target key node set, the path association mining process is performed on the disambiguated chemical product knowledge graph to obtain the entity indirect association information set. The disambiguated chemical knowledge graph is used to predict the link relationships to obtain a predicted relationship information set. Using an association rule engine, association rule mining is performed on the disambiguated chemical product knowledge graph to obtain a set of inference association information. The indirect association information set of entities, the predicted association information set, and the inferred association information set are deduplicated to obtain a deduplicated association information set, which serves as the multimodal association information set.

7. A chemical substance early warning and interception device, comprising: An identification unit is configured to identify the target chemical detection terminal set in response to the detection of a data connection request information set of the target chemical detection terminal set, and obtain a detection terminal identification information set. The chemical substance status information extraction unit is configured to control at least one target chemical substance detection terminal identified by the detection terminal identification information set, and to extract chemical substance status information from the multimodal target chemical substance information set corresponding to the at least one target chemical substance detection terminal to obtain a multimodal chemical substance status information set. The data deduplication unit is configured to perform data deduplication processing on the multimodal state information set of the chemical articles to obtain a deduplicated multimodal chemical article information set; The multimodal association identification unit is configured to perform multimodal association identification on the deduplicated multimodal chemical information set to obtain a multimodal association information set; The chemical risk identification unit is configured to perform chemical risk identification on the deduplicated multimodal chemical information set based on the multimodal association information set, to obtain the chemical information set to be tracked; The early warning and interception unit is configured to perform trajectory tracking and early warning on the set of chemical substances to be tracked, obtain a set of chemical substance trajectory early warning information, and control a set of chemical substance early warning devices and a set of chemical substance interception devices to perform early warning and interception on the set of chemical substances to be tracked corresponding to the set of chemical substances to be tracked.

8. An electronic device, comprising: One or more processors; Storage device, on which one or more programs are stored, When the one or more programs are executed by the one or more processors, the one or more processors implement the method as described in any one of claims 1-6.

9. A computer-readable medium having a computer program stored thereon, wherein, When the computer program is executed by a processor, it implements the method as described in any one of claims 1-6.

Images