504 results about "Event recognition" patented technology

The invention provides an intelligent event identification method and system based on a high-speed camera, and the method comprises the steps: setting the collection parameters of the high-speed camera, and triggering the camera to collect a target scene video stream. And hardware acceleration decoding processing is carried out on the collected original video data stream, real-time environment illumination information of the environment illumination sensor is obtained, and dynamic brightness equalization processing is executed. And performing motion adaptive denoising processing on the video sequence. Geometric distortion correction is carried out on the image sequence through camera calibration parameters, sub-pixel-level displacement vectors and dense optical flow field data of a moving object are extracted, and multi-scale morphological features are extracted. And the features are fused to generate motion feature data, the data are processed through a spatio-temporal joint event classification model, an event identification result is output, the result is bound with a high-precision timestamp, and event identification information is output to an industrial control system display device in real time. According to the invention, the accuracy and real-time performance of event identification can be improved.

The invention provides a hydro meteorology early warning method for an offshore oil and gas platform, and belongs to the technical field of offshore hydro meteorology. Extreme weather events are identified by adopting minimum probability abnormal event identification vectors to match abnormal characteristic parameters, and abnormal signal characteristic parameters are input into an ocean dynamics prediction model to calculate real-time sea condition parameters; calling a multi-temporal-spatial-scale early warning fusion matrix to combine with a wavelet decomposition technology and a recurrent neural network to realize multi-scale information integration, analyzing an environmental parameter change trend through a sea condition jump identification model and triggering an emergency response, dynamically adjusting system parameters according to a stability evaluation index vector, and optimizing prediction precision by adopting an early warning residual value compensation matrix. And finally, multi-level early warning information is generated and a real-time early warning notification is sent to an operator, so that the technical problem of insufficient early warning precision of an offshore oil and gas platform hydro meteorology early warning system in multi-spatio-temporal scale data fusion processing is solved.

The invention relates to a method and system for identifying a highway event by using a video large model, and the method comprises the steps: employing a three-stage processing architecture, firstly carrying out the real-time target detection and preliminary event judgment of a highway monitoring video stream through employing a YOLO algorithm, and generating an event candidate set; inputting the candidate events and the video clips thereof into a specially trained visual large model for deep semantic analysis and secondary reasoning; and finally, a reasoning result is rechecked through a rule engine, and false alarms are filtered by applying illusion suppression and a space-time association rule. According to the method, the real-time performance of traditional target detection and the deep reasoning capability of a visual large model are fused, so that the problems of high false alarm rate and high missing report rate of a traditional method are effectively solved, the accuracy and reliability of event identification in a complex traffic scene are remarkably improved, and meanwhile, the real-time processing capability of a system on multiple paths of high-definition video streams is ensured.

The invention discloses a method, a system and equipment for voice interaction inside and outside a vehicle and a storage medium, and relates to the technical field of intelligent cabins and human-computer interaction. Identifying an interaction trigger type (in-vehicle active request, out-of-vehicle passive request, or system active trigger) based on the perceived data and / or the vehicle event; determining a corresponding response permission strategy according to the trigger type, and determining whether to allow the system to respond based on the response permission strategy; and if so, calling a large language model to generate a voice text, synthesizing voice by combining external perception characteristics, and broadcasting the voice through a loudspeaker outside the vehicle or a sound box in the vehicle. According to the scheme, three types of interaction intentions including the in-vehicle active request, the out-vehicle passive request or the system active triggering are recognized, the large language model is called to generate the scene-adaptive voice text, and personalized voice synthesis and broadcasting are performed in combination with the external perception characteristics, so that the response efficiency, the expression naturalness and the object adaptability of the in-vehicle and out-vehicle voice interaction are improved.

The invention provides a wireless sensing platform with an edge computing function, and relates to the technical field of wireless sensors. In order to solve the problems that in the prior art, a wireless sensor is large in data redundancy, high in transmission delay and weak in emergency recognition and response capacity in a complex scene, a joint processing mechanism combining a causal event recognition assembly, a causal data fusion assembly, an event separation assembly and a data distribution assembly is adopted. By detecting a causal change event in an environment and dynamically extracting and fusing sensor data with a strong causal relationship, redundant data transmission caused by emergencies is reduced, transmission delay is reduced, the robustness of causal data fusion in a concurrent event scene is improved, and the reliability of data fusion is improved. According to the invention, efficient and accurate sensing and response to emergencies in a complex environment are realized.

The invention relates to the technical field of traffic event identification, and discloses an abnormal traffic event identification method and system based on a traffic large model, and the method comprises the steps: obtaining a traffic data flow, extracting an abnormal feature vector, and obtaining an abnormal signal candidate set; grouping the candidate sets and calculating a deviation degree, and if the deviation degree exceeds a threshold value, taking the deviation degree as a risk signal to form an input subset; environment variables are extracted from the subsets, a mapping relation is established, and anomaly recognition embedding representation is obtained; classifying the embedded representation, judging a congestion precursor and generating an early warning signal to obtain an early warning signal sequence; matching the sequence to obtain an abnormal event chain; if the integrity is higher than a threshold value, analyzing the type to obtain an abnormal event type; extracting a correlation feature vector from the type, pushing the correlation feature vector to a traffic management platform to obtain an instruction, and obtaining an emergency response trigger instruction sequence; and executing the instruction sequence to extract a feedback data stream, inputting the traffic large model to judge the accuracy rate, and if the judgment accuracy rate is met, determining an optimized anomaly recognition framework. The method can solve the problem of insufficient early warning capability.

The invention relates to the technical field of medical treatment, and discloses a knee joint knee protection treatment personalized planning system based on multi-modal data fusion. A data acquisition and preprocessing module of the system acquires a multi-modal data stream of the knee joint of a patient in real time and generates a standardized data packet; the treatment event identification module is used for automatically identifying and verifying key treatment events from the standardized data packet and generating a treatment event sequence; the historical treatment path query module queries a distributed database and extracts treatment modes of similar patients; the graph matching and path generation module is used for calculating the similarity between the current treatment event sequence and a historical treatment mode, generating a fusion treatment path when the similarity exceeds a threshold value and storing the fusion treatment path to the block chain network; the abnormity monitoring and analysis module monitors abnormal event points in the fusion treatment path in real time, triggers a path decomposition mechanism and performs multi-dimensional performance analysis; and the personalized scheme generation module is used for dynamically adjusting treatment parameters and generating a personalized treatment planning scheme.

The invention belongs to the technical field of image recognition, and particularly relates to a sheep manure conveying blockage recognition method and system based on machine vision, and the method comprises the steps: firstly fusing the local brightness deviation and gradient features of pixels, and calculating a visual interference index to generate a stability mask; a visual interference area caused by highlight, shadow and the like on the surface of the material is shielded; then, space gating is carried out on an inter-frame difference result of the image by using the mask, a robust motion feature map is generated, and a static connected region is extracted from the robust motion feature map; and finally, calculating a blockage index by combining the area of the static region and the average static degree, and comparing with a decision threshold to judge a blockage event and generate a response signal. According to the method, the visual interference degree under the complex working condition can be reduced, and the accuracy and the automation degree of conveying system blockage event recognition are improved.

The invention relates to the technical field of smart traffic, and discloses a smart traffic monitoring system and method based on multi-source data fusion, and the system comprises a multi-source data collection module, a spatial-temporal feature fusion engine, a hierarchical decision core, a decision execution module, and an online learning module. The method corresponds to the system. According to the method, heterogeneous traffic data are unified under the same time-space reference through a multi-modal fusion technology, and comprehensive and accurate road network state feature representation is constructed; the hierarchical decision-making core forms a closed-loop decision-making link from event identification to control strategy generation through close cooperation of an event sensing layer and a regional optimization layer, and realizes rapid response and accurate control of traffic abnormity; on-line learning continuously optimizes decision logic based on complete historical operation data so as to adapt to a continuously changing traffic environment; finally, the accuracy of traffic state perception and the timeliness of decision response are improved, and reliable technical guarantee is provided for efficient management and control of intelligent traffic.

The invention relates to a two-stage road traffic abnormal event identification method and system based on a visual large model, and the method comprises the steps: collecting the monitoring image and video data of an expressway and an urban expressway, and building a static image semantic data set and a dynamic video traffic semantic data set; utilizing the static image semantic data set to train a visual large model to obtain a first visual large model; constructing a same-preference data pair, and performing direct preference optimization training of the first visual large model by using the same-preference data pair to obtain a second visual large model; intercepting an abnormal video key frame based on the dynamic video traffic semantic data set, and performing parameter fine tuning on the second visual large model based on the abnormal video key frame to obtain a road traffic abnormal event recognition model; and collecting a monitoring image or video sequence in real time, and performing abnormal event identification by using the road traffic abnormal event identification model. Compared with the prior art, the traffic abnormal event identification method provided by the invention can effectively combine dynamic and static characteristics of data and is efficient.

The invention discloses a partition load growth situation prediction method and system oriented to extreme weather, and the method comprises the steps: firstly obtaining and preprocessing multi-source historical data, and building an extreme weather event recognition mechanism, so as to generate an extreme weather event tag sequence; then, the load response characteristics of all the partitions are quantified based on the extreme weather time period, and the partitions are divided into a plurality of load increase situation categories with different response modes by adopting a clustering algorithm; thirdly, independently constructing and training an exclusive load prediction model for each load growth situation category; and finally, during prediction execution, selecting the sentry subareas in each category, and dynamically calibrating original prediction results of other follower subareas in the same category by monitoring prediction residual errors of the sentry subareas in real time and calculating a prospective correction amount according to a deviation propagation model. According to the invention, combination of classified exclusive modeling and real-time dynamic correction is realized, and the precision and reliability of load prediction in extreme weather are improved.

The invention discloses an abnormal event attribution method and system based on LLM deep semantic feature extraction, and the method comprises the steps: obtaining a text data set in a to-be-analyzed time window, and each piece of text data is unstructured data related to a preset scene; aiming at each piece of text data, utilizing a large language model to extract a topic type and a property label of the text data, then performing data statistics on the topic types and the property labels corresponding to all the text data in the to-be-analyzed time window, and generating a structured feature vector; inputting the structured feature vector corresponding to the to-be-analyzed time window into a trained abnormal event recognition model, and judging whether an abnormal event occurs in the preset scene in the to-be-analyzed time window or not; and if it is judged that the abnormal event occurs, the feature with the highest contribution degree is judged through the SHAP value of each feature in the structured feature vector, and corresponding text data is positioned based on the feature. According to the method, the sudden abnormal events of the system can be identified and subjected to root tracing.

The invention discloses a dynamic environment and multi-event collaborative awareness method based on a multi-modal neural network. According to the method, multi-modal data is obtained through a multi-source sensing device, and a multi-modal data sequence under a unified reference coordinate is formed through time synchronization and space registration. And inputting the data sequence into a multi-modal deep neural network, and adaptively correcting each modal feature weight according to a confidence coefficient change rate between time slices to obtain corrected feature information. And performing external target identification and internal event identification on the feature information to generate a candidate event set, establishing an event priority relationship, and outputting a collaborative event result including an event tag, a spatial position and a time range. And finally, the risk level is determined according to a collaborative event result, and a corresponding early warning or control instruction is generated, so that intelligent identification and collaborative decision-making of multiple events in a complex dynamic environment are realized, and the real-time performance and accuracy of the system in multi-source data fusion and safety monitoring are improved.

According to the abnormal event identification method based on the multi-modal model provided by the invention, multi-modal data capable of comprehensively reflecting states and environment information of old people is obtained by fusing various modal data such as the video data, the audio data, the environment data and the user related data, the limitation of single-modal data is reduced, and the identification efficiency is improved. And misjudgment and missed judgment caused by single modal data are reduced, so that the accuracy and reliability of abnormal event identification are improved. And furthermore, through synchronous acquisition and alignment of multi-modal data, the problem of time sequence dislocation caused by data source dispersion is solved, and the accuracy of abnormal event identification is further improved. Furthermore, different fusion strategies are adopted based on different detection time types, so that the internal relation between modes can be better mined, and the abnormal event recognition precision under various different application scenes and event types is improved. The method can be applied to the financial field and the medical field, safety monitoring is carried out on the accident safety of the user, and the recognition accuracy of the abnormal event is improved.

The invention belongs to the technical field of robot control, and particularly discloses a wire harness assembly robot control method and system based on machine learning, and the method comprises the steps: firstly collecting the joint torque, the position error of an end effector, the posture change and other multi-dimensional sensor data in real time when a robot wire harness is plugged; intercepting and constructing a multi-channel contact event tensor by a sliding time window; inputting the effective contact event into a pre-training sparse auto-encoder, encoding to obtain a sparse vector, reconstructing a reconstructed tensor, and decomposing a sparse component of the effective contact event and a residual component of interference noise; identifying an effective contact event by judging whether the norm of the sparse vector L1 exceeds a first threshold value or not; and finally, calling a predefined fine tuning control strategy to execute an action response. Weak contact event recognition precision and action response accuracy are improved, assembly quality and safety are guaranteed, and the method is suitable for efficient control of the wire harness assembly robot.

The invention belongs to the technical field of communication, and provides a communication optical cable state real-time online monitoring method and system. Calculating the vibration displacement of the optical cable based on the original back scattering light signal, and calculating the real strain and temperature of the optical cable based on a Brillouin scattering signal and a Raman scattering signal obtained by separating the original back scattering light signal; carrying out filtering, denoising, normalization and time-space alignment processing on the vibration displacement of the optical cable, the real strain of the optical cable and the temperature, and binding the vibration displacement and the real strain with the obtained geographic coordinates; obtaining time domain, frequency domain and time-frequency domain features based on the preprocessed vibration displacement, real strain and temperature data to form a multi-dimensional feature vector, inputting the feature vector into a pre-trained event recognition model, and outputting an abnormal event classification result; and judging and triggering abnormal event early warning based on an output result of the event recognition model and a preset dynamic threshold.

The invention provides a submarine earthquake optical fiber monitoring data processing method and device, equipment and a medium, which can be applied to the technical field of communication. The method comprises the following steps: acquiring original optical signal data generated by a submarine seismic signal; preprocessing the original optical signal data to obtain initial monitoring data; performing seismic event identification on the initial monitoring data to obtain seismic event data; and compressing the seismic event data to obtain compressed monitoring data. According to the method provided by the invention, a series of data processing such as preprocessing, filtering and denoising, seismic event identification and compression processing are carried out on the optical fiber monitoring data in sequence, so that the efficiency and precision of accessing the optical fiber monitoring data to the seismic monitoring platform are improved.

The invention discloses a highway emergency disposal method and system, and belongs to the technical field of emergency disposal. The method comprises the following steps: an event identification stage: deploying multi-source sensing fusion equipment on a highway to automatically detect emergencies, and increasing deployment density in a tunnel and an accident high-incidence road section; an event confirmation stage: performing multi-dimensional fusion analysis and triple verification based on the multi-source sensing data detected by the multi-source sensing fusion equipment; in the scheme triggering stage, according to the confirmed event information, an equipment linkage control strategy is generated based on the multi-level decision-making system; an equipment linkage control stage: performing emergency handling according to the generated equipment linkage control strategy; and a state feedback and optimization stage: carrying out self-learning and iterative optimization on the equipment linkage control strategy based on a reinforcement learning optimization framework. Compared with a traditional manual disposal mode, the system has the advantages that the response speed is obviously increased, the secondary accident rate is greatly reduced, and the energy consumption of equipment is reasonably optimized.

The micro-service processing system provided by the embodiment of the invention comprises an event identification module which is used for defining a business domain event type and registering an event attribute structure; the event bus module constructs a distributed communication network containing a dual-channel message queue, and the event bus module implements partition routing according to event types; the event processing module is composed of an event producer and a consumer which are executed asynchronously, the event producer packages service actions into a standardized event data packet, and the consumer processes subscription events through a thread pool; the event storage module adopts a hierarchical log database for persistence of an event sequence and supports reestablishment of a service state according to a timestamp; and the security control module is used for implementing service authentication and event content encryption on a transmission layer. Through event bus dual-channel design, sender box transaction binding and hierarchical storage optimization, millisecond-level event processing delay is realized while the reliability of distributed transactions is ensured, and second-level state recovery can be realized based on a complete event log when a service fault occurs.

The invention discloses an intelligent public opinion research and judgment method and system for public security actual combat, and the method achieves the high-throughput data capture and real-time public opinion inflow through multi-source crawler, breakpoint resume, incremental collection and asynchronous concurrent processing, and greatly reduces the workload of manual screening. A unified, quantifiable and structured public opinion risk grading standard is established, and risk grading standard unification is realized by constructing three types of structured public opinion dictionaries of event subjects, event properties and sensitivity grades; the grading rule is quantified and objective and adapts to the grading system of the public security actual combat, and the subjectivity and inconsistency of research and judgment are remarkably reduced; and the recessive public opinion recognition capability is improved through the deep semantic model. According to the method, the BERT model, keyword vector fusion and the police affair large model are combined, so that accurate analysis of complex semantics such as metaphor, black talk and irony mood is realized; high-robustness semantic matching and duplicate removal are carried out; through accurate extraction of risk factors and event identification, the problem of high missed judgment rate in the prior art is improved.

The invention discloses a digital twinning-oriented animation linkage method and device, equipment and a storage medium. The method comprises the steps that based on classes in a domain ontology model and attributes of the classes, operation data of the Internet of Things system are converted into first semantic data, and the domain ontology model comprises an environment class, an equipment class and a behavior class; performing event identification on the first semantic data to obtain a first event; reasoning the first event based on a preset reasoning rule set to obtain a first reasoning result, the first reasoning result comprising a first response behavior for the first event; and controlling a digital twinborn body corresponding to the Internet of Things system to execute an animation matched with the first response behavior. Therefore, a closed-loop control chain of data driving, semantic recognition, event reasoning and animation linkage is formed, high reliability is achieved, complex dynamic linkage is supported, and therefore intelligent, automatic and visual expression of the system is effectively improved.

The invention discloses a hydro-junction construction period structure safety assessment method and system based on digital twinning, and relates to the technical field of hydro-junction safety assessment, and the method comprises the steps: carrying out the safety correlation analysis of a hydro-junction construction period structure; constructing a multi-level digital twinborn space; the method comprises the following steps: collecting construction monitoring data of a hydro-junction during a construction period, synchronously mapping the construction monitoring data to a multi-level digital twin space for level simulation, and identifying a first type of safety events; and when the first type of security events exist, carrying out influence range tracing and risk conduction analysis, and generating a structure security assessment conclusion. According to the invention, the technical problem of insufficient evaluation comprehensiveness and accuracy caused by the fact that dominant safety event identification and hidden risk mining are difficult to consider in the construction period structure safety evaluation of the hydro-junction in the prior art is solved, and accurate monitoring and targeted risk research and judgment of the construction period structure safety are realized; and the technical effect of improving the comprehensiveness and accuracy of structure safety evaluation in the construction period is achieved.

Information about power events on the electrical grid may be determined by processing reports of power anomalies from power monitors installed at various points on the electrical grid, such as in buildings of end users of electrical power. A power monitor may process sensor measurements of the power line to determine that a power anomaly has occurred. The power monitor may transmit information about the power anomaly, such as the time and location, to a processing location. The processing location may select power anomaly reports having a similar time and location to determine information about a power event that caused the power anomalies. The information about the power event may include the type, time, and location of the power event. A notification may be sent about the power event to facilitate repairs and reduce risks, such as the risk of electrical fires.

The invention discloses a lightweight optical fiber vibration intrusion event identification method and system for perimeter security and protection, and belongs to the technical field of optical fiber sensing technology and mode identification, and the method comprises the steps: collecting an original vibration signal of a perimeter monitoring region through a distributed optical fiber vibration sensing system; performing wavelet threshold de-noising preprocessing on the original vibration signal to obtain a de-noised signal; performing feature extraction on the denoised signal, and constructing a high-dimensional feature vector; performing dimension reduction processing on the high-dimensional feature vector by using a linear discriminant analysis method to obtain a low-dimensional classification feature vector; and inputting the low-dimensional classification feature vector into a pre-trained lightweight convolutional neural network model for classification and identification, and outputting a corresponding intrusion event category. According to the method, through cooperation of front-end LDA dimension reduction and a rear-end lightweight network, the model parameter quantity and calculation overhead are greatly reduced while high recognition precision is guaranteed, efficient real-time deployment on edge equipment is achieved, and the method is suitable for intrusion detection in perimeter security and protection.

The invention discloses a wireless communication transmission efficiency optimization method and system based on data analysis, and relates to the technical field of wireless communication, and the method comprises the following steps: S1, constructing a local link fluctuation characteristic analysis model; s2, constructing a sudden interference event identification criterion model based on a threshold value; s3, determining whether to execute channel bypass switching or not according to the criterion result in real time; s4, updating the sequence of the service buffer queue in real time; s5, constructing a transmission queue mechanism based on priority; and S6, carrying out self-adaptive fine tuning optimization on the threshold value in the step S2. According to the method, a dynamic optimization system including link fluctuation perception, event identification, channel switching, resource allocation and adaptive feedback is constructed, so that the response speed and the identification precision of the system to burst interference are improved, and the adaptive capacity of a network scheduling strategy to environmental change is enhanced; and finally, the communication transmission efficiency is remarkably improved, and the key service is stably guaranteed.

The invention relates to the technical field of intelligent traffic, in particular to a road traffic incident intelligent identification and filtering method and system, and the method comprises the steps: collecting video image data, radar point cloud data and environment sensing data, and generating the track information of a traffic target; performing space-time alignment and feature fusion on the trajectory information and the radar point cloud data to form multi-modal features; the method comprises the following steps: preliminarily identifying a traffic abnormal event based on multi-modal feature analysis, and generating event report information comprising an event image and associated features; performing deep semantic understanding on the event report information, analyzing the event image, the track information and the environment sensing data according to the associated features, and determining the semantic category of the traffic abnormal event; executing intelligent filtering processing based on the semantic category; and recording the traffic abnormality subjected to intelligent filtering processing as an effective traffic event, and generating a traffic recognition result. According to the method, deep semantic understanding is carried out through rich multi-modal features, and the overall accuracy and processing efficiency of traffic incident recognition are improved.

The invention relates to the technical field of AI intelligent calculation, in particular to a broadcast content real-time analysis interaction method and system based on an AI multi-modal large model, and the method comprises the steps: achieving the precise event recognition and structural description through the efficient modeling of single-modal real-time data, and providing the standardized input for the subsequent content generation; then, broadcast texts meeting object features and scene requirements are generated based on event information and global static audience feature configuration, and quantified semantic priorities are calculated to support strategy decision; on the basis, performing deep intention recognition on the generated content by utilizing multi-modal large model reasoning, introducing targeted parameters such as historical broadcast diversity punishment and interference sensitivity adjustment, and generating an executable scheduling strategy; and finally, through area matching and terminal capability verification, mapping the strategy parameter into a specific equipment control instruction, and issuing and executing the specific equipment control instruction to form a closed loop from content generation to precise broadcasting.

The invention discloses a table tennis motion detection system based on pure visual perception. The table tennis motion detection system comprises a table tennis detection and event monitoring module, a human body-racket posture estimation module and a time sequence action recognition and quality evaluation module. The table tennis ball detection and event monitoring module is based on an improved DTTNet model and realizes table tennis ball track detection and event identification such as table touching, net touching and net passing through through multi-frame video stack and deformable convolution; the human body-racket posture estimation module adopts a lightweight detection network and high-resolution posture estimation network combined reasoning structure, outputs human body key point and racket three key point coordinates, and constructs a human body-racket combined kinematics model; the time sequence action recognition and quality evaluation module fuses multi-view video and posture features, classification and quality evaluation of table tennis technical actions are achieved through time sequence boundary matching and multi-scale feature analysis, the table tennis technical action recognition and quality evaluation system has the advantages of being high in detection precision and high in real-time performance, and automatic recognition and quantitative evaluation of table tennis movement can be achieved.

The invention relates to the technical field of pipeline safety monitoring, in particular to a long-distance pipeline risk event distinguishing monitoring system and method based on artificial intelligence. The artificial intelligence-based long-distance pipeline risk event distinguishing monitoring method comprises the following steps of S10, acquiring humidity time sequence data and anode output current time sequence data of soil in an area where a target pipe section is located, and respectively drawing a humidity time curve and a current time curve; acquiring a vibration signal acquired by the optical fiber sensing network; and S20, periodically fluctuating according to the current-time curve. According to the method, the relation between the regional self-adaptive humidity saturation interval and the corrosion rate is established, the huge cost for customizing a vibration analysis model for each region is avoided, the event recognition accuracy is improved under high-risk scenes such as coating damage by utilizing deep coupling of the corrosion state and vibration, and the event recognition efficiency is improved. And passive alarm is converted into active and accurate early warning based on humidity prediction.

The invention discloses an implementation method and system of a low-power-consumption and high-reliability built-in automobile data recorder, and relates to the technical field of vehicle-mounted electronics and intelligent monitoring. Automatically waking up the main control module and initializing a sensor, a lens, a memory and an accelerometer; constructing a multi-modal data stream, storing historical data of the latest T1 seconds through a cache region, and performing beforehand pre-recording; continuously analyzing the running state and the acceleration abnormal value of the vehicle, and when the multi-source data meets a set threshold value, judging as event triggering; when an event trigger signal is detected, combining the pre-recorded data in the cache region with a video T2 seconds after an event occurs, and executing a hierarchical locking strategy; and an uploading strategy is automatically selected according to the wireless network bandwidth and the event level. According to the method, the multi-modal data stream cache region is constructed, the latest vehicle video and sensing data are stored in real time, the evidence obtaining effectiveness is improved, and the event recognition precision is improved.