Emergency response method and device based on vehicle sentry mode, equipment and medium

Abstract

The invention provides an emergency response method, device and equipment based on a vehicle sentry mode and a medium. The method comprises the following steps: acquiring video, audio and thermal radiation data of a vehicle surrounding environment through a multi-mode sensor; inputting the data into a preset behavior recognition model, recognizing an abnormal event type and a corresponding event level, dynamically intercepting evidence data according to the event type and the event level, encrypting the evidence data through a preset encryption algorithm, and transmitting the encrypted evidence data to a cloud for storage; according to the method, edge calculation and cloud collaborative analysis are combined, real-time intelligent recognition, graded response and evidence chain generation of emergencies are achieved, the false alarm rate is effectively reduced, the emergency rescue efficiency is improved, and meanwhile data privacy and tamper-proofing performance are guaranteed.

Description

Technical Field

[0001] This invention belongs to the field of intelligent connected vehicle technology, and in particular relates to an emergency response method, device and medium based on vehicle sentry mode. Background Technology

[0002] Current vehicle sentry mode primarily monitors the vehicle's surroundings using onboard cameras and sensors. However, its functionality is limited to video recording and basic anti-theft alarms, lacking the ability to intelligently assess and proactively respond to complex emergencies. Existing technologies rely solely on cameras to passively record images, failing to analyze human behavior or environmental anomalies (such as fights or fires) in real time, leading to missed or delayed reporting of critical events. Some solutions attempt to incorporate motion detection technology, but these suffer from high false alarm rates, reliance on manual video review, and an inability to differentiate hazard levels or trigger coordinated rescue efforts.

[0003] Therefore, there is an urgent need for a better vehicle emergency response method to improve the efficiency of handling public safety incidents and the ability to collect evidence in the judicial system. Summary of the Invention

[0004] This invention provides an emergency response method, device, and medium based on vehicle sentry mode to solve the existing technical problems of high false alarm rate, inability to distinguish danger level, or inability to trigger coordinated rescue.

[0005] This application provides an emergency response method based on vehicle sentry mode. The method includes: collecting environmental data of the vehicle's location, the environmental data including at least video data, audio data, and thermal radiation data; inputting the environmental data into a preset behavior recognition model to obtain abnormal event types and corresponding event levels, the behavior recognition model being trained on scenarios divided by the environmental data; extracting corresponding evidence data from the environmental data according to the abnormal event type and the event level, and encrypting the evidence data; and storing the encrypted evidence data in a preset storage center.

[0006] In one embodiment of the present invention, obtaining the abnormal event type and the corresponding event level includes: extracting multi-dimensional features from the collected environmental data to generate a spatiotemporal fusion feature vector; inputting the spatiotemporal fusion feature vector into a preset behavior recognition model for event probability analysis to obtain a confidence parameter corresponding to the abnormal event type; when the confidence parameter of the abnormal event type exceeds a preset threshold and the duration exceeds a preset duration threshold, it is determined that the abnormal event type has been triggered; based on the abnormal event type, the duration of the continuously triggered abnormal event is monitored, and the event level is determined according to the duration.

[0007] In one embodiment of the present invention, extracting corresponding evidence data from environmental data includes: determining the start and end time points of the data segment to be extracted based on a preset time window parameter matched with the abnormal event type and the event level; extracting continuous data segments within a target time interval from the continuously collected environmental data stream based on the start and end time points; adding an anti-tampering identification layer to each data unit in the continuous data segment, wherein the anti-tampering identification layer includes a watermark superimposed with a data generation time marker, a spatial location marker, and a device identity marker; and integrating the continuous data segments with the added anti-tampering identification layer into a structured evidence data package according to the time sequence, wherein the time span of the data segment is positively correlated with the event level.

[0008] In one embodiment of the present invention, encrypting the evidence data includes: generating an encryption key uniquely bound to the evidence data and storing the encryption key in a protected independent area; dividing the evidence data into multiple data units and performing iterative encryption operations on each data unit based on preset encryption logic; adding an integrity verification identifier to the encrypted data units to form a tamper-proof evidence package.

[0009] In one embodiment of the present invention, storing encrypted evidence data in a preset storage center includes: mapping the tamper-proof evidence package to an independent storage partition of the preset storage center according to the event level; constructing a redundant storage topology containing a primary storage node and at least one backup storage node, and synchronously writing the tamper-proof evidence package to the primary storage node and the backup storage node; generating a multi-node association index for the tamper-proof evidence package based on the redundant storage topology, the multi-node association index containing an event level label and a timestamp sequence; performing periodic data verification according to the redundant storage topology, and triggering a data replacement process from the backup storage node to the primary storage node when the verification of the tamper-proof evidence package of the primary storage node is abnormal; and setting access permission rules for the storage partition based on the event level to restrict unauthorized operation commands from accessing the tamper-proof evidence package.

[0010] In one embodiment of the present invention, before encrypting the evidence data, the method further includes: scanning dynamic image frames in the evidence data based on a preset sensitive information feature library, identifying privacy-sensitive areas containing facial features, vehicle identification features, or biological behavioral features, and retaining an encrypted backup of the original data to a preset independent security partition; blurring the identified privacy-sensitive areas to generate privacy-desensitized data with masked tags; establishing a mapping relationship between privacy masked tags and event levels, and dynamically adjusting the visualization permissions of the masked areas according to the event levels; encapsulating the privacy-desensitized data and the mapping relationship into a privacy-enhanced evidence unit, and synchronously embedding the mapping relationship as independent metadata during the encryption process; generating a privacy-desensitization operation log, recording the masked area location, operation time, and event level association information, and synchronously storing the privacy-desensitization operation log to a preset monitoring node.

[0011] This application provides an emergency response device based on vehicle sentry mode. The device includes: a multimodal sensing module for collecting environmental data around the vehicle; an edge computing module for generating a spatiotemporal fusion feature vector of the environmental data and determining the abnormal event type and event level based on the spatiotemporal fusion feature vector, wherein the behavior recognition model is trained through scenarios divided by the environmental data; a response execution module for triggering a hierarchical response protocol according to the abnormal event type and the event level; an evidence chain management module for capturing environmental data according to a time window matched by the event level and encrypting the evidence data based on a preset encryption algorithm; and a hierarchical storage module for retaining the corresponding encrypted evidence data according to the event level.

[0012] In one embodiment of the present invention, the evidence chain management module includes: an anti-tampering identification unit, used to overlay a watermark layer containing time markers, spatial location markers, and device identification markers onto the intercepted environmental data; a privacy desensitization unit, used to scan privacy-sensitive areas containing facial features, vehicle identification features, or biological behavioral features in dynamic image frames based on a feature library; a masking processing subunit, used to blur the information in the privacy-sensitive areas and generate corresponding privacy-desensitized data; and an original data backup subunit, used to encrypt and store the undesensitized data in the privacy-sensitive areas to a preset independent secure partition.

[0013] This application provides an electronic device, the electronic device comprising: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the emergency response method based on vehicle sentry mode as described above.

[0014] This application provides a computer-readable storage medium, characterized in that it stores a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the emergency response method based on vehicle sentry mode as described above.

[0015] The beneficial effects of this invention: This invention proposes an emergency response method, device, equipment, and medium based on a vehicle sentry mode. This method significantly improves the identification and handling efficiency of emergencies by collecting environmental data and combining it with a behavior recognition model to analyze abnormal events. First, the integration of data from multiple sensors overcomes the limitations of single monitoring methods, enabling more accurate identification of complex events such as fires, fights, and people falling to the ground, effectively reducing false alarms and missed alarms. Second, it automatically extracts key evidence data based on event level, ensuring the integrity required for judicial evidence collection while avoiding redundant data occupying storage resources. Simultaneously, encryption technology ensures the confidentiality and tamper-proof nature of evidence data, meeting legal compliance requirements. Furthermore, the encrypted data is centrally stored in a pre-set storage center, achieving structured management and facilitating rapid retrieval and review analysis, significantly improving event handling efficiency. Overall, the above technical measures collectively solve the problems of reliance on manual analysis, delayed response, and incomplete evidence in the traditional sentry mode, providing reliable support for the intelligent handling of public safety incidents. Attached Figure Description

[0016] The accompanying drawings, which are incorporated in and form part of this specification, illustrate embodiments consistent with this application and, together with the description, serve to explain the principles of this application. It is obvious that the drawings described below are merely some embodiments of this application, and those skilled in the art can obtain other drawings based on these drawings without any inventive effort.

[0017] In the attached diagram: Figure 1 This is a schematic diagram illustrating the implementation environment of an emergency response method based on vehicle sentry mode according to an embodiment of the present invention. Figure 2 This is a flowchart of an emergency response method based on vehicle sentry mode provided in one embodiment of the present invention; Figure 3 This is a schematic diagram illustrating the complete steps of an emergency response method based on vehicle sentry mode provided in one embodiment of the present invention; Figure 4 This is a schematic diagram of the structure of an emergency response device based on vehicle sentry mode provided in one embodiment of the present invention; Figure 5 A schematic diagram of the structure of a computer system suitable for implementing the electronic device of the present application is shown. Detailed Implementation

[0018] The following specific examples illustrate the implementation of the present invention. Those skilled in the art can easily understand other advantages and effects of the present invention from the content disclosed in this specification. The present invention can also be implemented or applied through other different specific embodiments. Various details in this specification can also be modified or changed based on different viewpoints and applications without departing from the spirit of the present invention. In the absence of conflict, the following embodiments and features in the embodiments can be combined with each other.

[0019] It should be noted that the illustrations provided in the following embodiments are only schematic representations of the basic concept of the present invention. The drawings only show the components related to the present invention and are not drawn according to the actual number, shape and size of the components in the actual implementation. In the actual implementation, the form, quantity and proportion of each component can be arbitrarily changed, and the layout of the components may also be more complex.

[0020] In the following description, numerous details are explored to provide a more thorough explanation of embodiments of the invention. However, it will be apparent to those skilled in the art that embodiments of the invention may be practiced without these specific details. In other embodiments, well-known structures and devices are shown in block diagram form rather than in detail to avoid obscuring embodiments of the invention.

[0021] Please see Figure 1 , Figure 1 This is a schematic diagram illustrating the implementation environment of an emergency response method based on vehicle sentry mode, as provided in an embodiment of the present invention.

[0022] like Figure 1 As shown, the implementation environment of the emergency response method based on vehicle sentry mode consists of a data acquisition module 101 and a cloud data storage center 102. The two interact with each other through 5G-V2X or MQTT (Message Queuing Telemetry Transport) protocol to collaboratively complete the real-time identification, response and evidence management of abnormal events.

[0023] The data acquisition module 101 integrates multiple sensors, including a 4K camera, microphone array, infrared thermal imaging module, biosensor, GPS, and accelerometer, to collect video, audio, thermodynamic features, and location information in real time. The edge computing unit filters, performs analog-to-digital conversion, and extracts spatiotemporal features from the raw data, such as target displacement velocity, acoustic signature spectrum energy, and temperature gradient changes. After generating standardized feature vectors, the data is uploaded to the cloud via the vehicle communication protocol, and a local visualization interface is provided for the driver to monitor the status.

[0024] The cloud-based data storage center 102 serves as the core processing unit, encompassing a behavior recognition model engine, an event-level response module, a judicial evidence management platform, and multi-departmental collaboration interfaces. After receiving data from the vehicle, the cloud first performs cross-modal analysis of video, audio, and thermodynamic features using a spatiotemporal convolutional neural network, outputting an event probability matrix, such as the confidence levels for fire, fight, and falling to the ground. Based on preset thresholds, such as "temperature gradient abrupt change > 15℃ / second" or "fight confidence level > 85%", a differentiated response is triggered. This differentiated response includes at least directly connecting to 119 (police), dialing 110 (police), and calling 120 (ambulance).

[0025] In terms of evidence management, environmental data within the event time window is dynamically captured in the cloud. For example, for a Level 1 event, data from 30 minutes before the event to 10 minutes after the event is retained, overlaid with timestamps, GPS coordinates, and device ID watermarks to generate a structured evidence package. The data is encrypted using AES-256 and bound to a Hardware Security Module (HSM) key, and simultaneously written to redundant cloud storage nodes and local tamper-proof solid-state drives (SSDs), forming a closed loop for judicial evidence collection. At the same time, access permissions are set according to the event level; for example, fire departments have exclusive access to fire data, and privacy-de-identified operation logs are recorded via blockchain to ensure compliance.

[0026] The cloud platform also pushes event tags, such as "fire - abnormal high temperature," to the emergency command center through multi-departmental collaboration interfaces, enabling joint dispatching by departments such as fire, public security, and medical services. For example, in the event of a fire, the location data of the fire source can be synchronized; after a fight escalates, the police, in conjunction with traffic police, can pinpoint the location of vehicles.

[0027] This environment, through vehicle-cloud collaborative computing and multi-level security strategies, reduces the response efficiency of emergencies from traditional manual handling to milliseconds, providing an intelligent and compliant end-to-end solution for vehicle safety protection and judicial evidence collection.

[0028] like Figure 2 As shown, in an exemplary embodiment, the emergency response method based on vehicle sentry mode includes at least steps S210 to S240, which are described in detail below: Step S210: Collect environmental data of the vehicle's location. The environmental data includes at least video data collected by the camera, audio data collected by the microphone, and thermal radiation data collected by the infrared thermal imaging module.

[0029] In one embodiment of this application, a multimodal sensor array is used to collect real-time environmental data around the vehicle. Specifically, this includes: a monitoring system consisting of four 200° ultra-wide-angle cameras arranged in a ring continuously capturing 360° high-definition video data with 4K resolution and 60fps frame rate; a directional microphone array using beamforming technology to accurately collect environmental audio signals, filter background noise, and focus on sounds in specific directions, such as cries for help, explosions, or abnormal noises; and an infrared thermal imaging module integrated inside the windshield scanning the surrounding scene at a frequency of 30 frames per second to generate a thermal radiation distribution map ranging from -20°C to 550°C.

[0030] The aforementioned sensor data is aligned with the spatiotemporal dimensions through the feature fusion layer of the edge computing unit to extract dynamic target motion trajectory, sound source localization, and temperature gradient change features to form a multimodal perception data stream, thereby obtaining environmental data of the vehicle's location.

[0031] Step S220: Input the environmental data into the preset behavior recognition model to obtain the abnormal event type and the corresponding event level. The behavior recognition model is trained by dividing the scene from the environmental data.

[0032] In one embodiment of this application, environmental data collected by multimodal sensors is processed by the feature fusion layer of an edge computing unit to extract spatial-temporal feature vectors, such as the motion trajectory of video frames, voiceprint features of audio, and temperature gradient change rate. These feature vectors are then input into a pre-set behavior recognition model trained in the cloud. This model is trained based on a virtual scene library generated by Unity3D, covering simulation data of typical abnormal events such as fights, falls, and fires, and possesses multimodal perception capabilities. After obtaining the event probability matrix based on this behavior recognition model, a threshold determination mechanism is used to determine the type of abnormal event: a temperature gradient change exceeding 15℃ / second is classified as a Level 1 event (fire); a fight with a confidence level exceeding 85% is classified as a Level 2 event (fight); and a person falling for more than 3 minutes is classified as a Level 3 event (fall). It should be noted that Unity3D is a cross-platform comprehensive game engine and real-time 3D content creation tool.

[0033] In one embodiment of this application, obtaining the abnormal event type and the corresponding event level includes: extracting multi-dimensional features from the collected environmental data to generate a spatiotemporal fusion feature vector; inputting the spatiotemporal fusion feature vector into a preset behavior recognition model for event probability analysis to obtain a confidence parameter corresponding to the abnormal event type; when the confidence parameter of the abnormal event type exceeds a preset threshold and the duration exceeds a preset duration threshold, it is determined that an abnormal event type has been triggered; based on the abnormal event type, the duration of the continuously triggered abnormal event is monitored, and the event level is determined according to the duration.

[0034] In a specific embodiment of this application, the process of extracting multi-dimensional features from the collected environmental data and generating a spatiotemporal fusion feature vector is as follows: First, a 4K ultra-high-definition video stream is acquired through a vehicle-mounted circular camera array. Motion trajectory analysis algorithms are used to extract the displacement velocity and direction features of the target object in the video frames. Simultaneously, a directional microphone array is used with beamforming technology to separate environmental noise from the target audio signal, extracting the spectral energy distribution and sudden sound pressure change features from the voiceprint characteristics. The infrared thermal imaging module monitors the gradient changes of the environmental temperature field in real time, generating a spatial distribution map of the temperature change rate and thermal radiation intensity.

[0035] When the three types of heterogeneous data mentioned above are processed by the feature fusion layer of the edge computing unit, a spatiotemporal convolutional neural network is used to perform cross-modal correlation between the motion vectors of video frames and audio voiceprint features, constructing a three-dimensional feature vector that includes visual motion patterns, auditory event features, and thermodynamic anomalies. After this feature vector is input into a preset behavior recognition model, the model performs event probability analysis based on a virtual scene library generated by Unity3D (covering simulation data of typical abnormal events such as fights, fires, and people falling to the ground), and outputs a probability matrix including the confidence levels of fire, fights, and falls.

[0036] When the confidence level of a fire exceeds 80% and the duration of a sudden temperature gradient change reaches 2 seconds, the system determines that a Level 1 event is triggered; when the confidence level of a fight exceeds 85% and the cumulative number of frames of continuous physical conflict reaches 15, the system determines that a Level 2 event is triggered; when the confidence level of a fall exceeds 75% and the target object remains stationary for more than 3 minutes, the system determines that a Level 3 event is triggered.

[0037] Furthermore, for triggered abnormal event types, their status changes will be continuously monitored: if a fire event lasts for more than 10 minutes and the temperature gradient remains above 12℃ / second, the event level will be automatically upgraded from Level 1 to the highest priority; if a fight involves weapon waving and the confidence peak exceeds 92%, an emergency escalation response for Level 2 events will be triggered; for Level 3 events, when the heartbeat signal of a fallen person is detected as abnormal by the vehicle-mounted biosensor, the system will adjust the event level from Level 3 to a critical event requiring immediate rescue. It should be noted that in the scheme proposed in this embodiment, the behavior recognition model also dynamically optimizes feature weights through adversarial example training technology to avoid misjudgments caused by environmental interference such as low light at night and rain / fog obstruction. Simultaneously, it combines time series analysis algorithms to predict the trend of event development, ensuring the accuracy and real-time nature of event level determination. Among them, adversarial example training technology is a defensive training method that enhances the robustness of machine learning models by introducing adversarial examples. During the training process, the model is actively shown carefully designed adversarial examples, which are samples that cause the model to misclassify after adding small perturbations to the original input. This forces the model to learn more robust feature representations, thereby improving its ability to defend against real adversarial attacks.

[0038] Step S230: Based on the abnormal event type and event level, extract the corresponding evidence data from the environmental data and encrypt the evidence data.

[0039] In one embodiment of this application, when a Level 1 fire event is triggered, 4K ultra-high-definition video frame sequences from 30 minutes before and 10 minutes after the event are automatically captured, and the temperature gradient change curve recorded by the infrared thermal imaging module is extracted simultaneously; when a Level 1 brawl event is triggered, a mixed video and audio data stream from 15 minutes before and 5 minutes after the event is captured, including voiceprint features captured by the microphone array and limb movement trajectories recorded by the camera; when a Level 1 fall event is triggered, video segments from 1 minute before and 3 minutes after the event are captured, and vital sign data detected by the vehicle-mounted biosensor is superimposed. After the captured evidence data is processed by the edge computing unit, the video file is first divided into several data blocks using segmentation encryption technology and encrypted block by block using the AES-256 algorithm. The encryption key is dynamically generated by the Hardware Security Module (HSM) and stored in the vehicle security chip. Subsequently, the encrypted data packet is uploaded to the cloud evidence library via the TLS 1.3 protocol. The local synchronous execution protocol encryption operation overlays a digital watermark containing timestamps, geographic locations, and unique device numbers onto video frames. The watermark information is bound to the original data using hash chain technology to prevent tampering. The final encrypted evidence data is stored in the cloud for a period of three years for major events and one month after verification for minor events, and supports judicial institutions to retrieve and decrypt it through multi-level authorization authentication.

[0040] In one embodiment of this application, extracting corresponding evidence data from environmental data includes: matching preset time window parameters according to the type and level of the abnormal event to determine the start and end time points of the data segment to be extracted; extracting continuous data segments within a target time interval from the continuously collected environmental data stream based on the start and end time points; adding an anti-tampering identification layer to each data unit in the continuous data segment, the anti-tampering identification layer including a watermark superimposed with a data generation time marker, a spatial location marker, and a device identification marker; and integrating the continuous data segments with the added anti-tampering identification layer into a structured evidence data package according to the time sequence, wherein the time span of the data segment is positively correlated with the event level.

[0041] In one specific embodiment of this application, the specific implementation method for extracting corresponding evidence data from environmental data is as follows: First, based on the event level, a preset time window parameter is matched to determine the start and end times of the data segment to be extracted. For a Level 1 fire event, the time window is 30 minutes before the event to 10 minutes after; for a Level 1 brawl event, the time window is 15 minutes before the event to 5 minutes after; and for a Level 1 fall event, the time window is 1 minute before the event to 3 minutes after. Based on the determined start and end times, continuous data segments within the target time interval are extracted from the continuously collected environmental data stream. These data segments include video frame sequences, audio signals, and temperature change curves recorded by multimodal sensors.

[0042] It should be noted that for the same type of abnormal event, the corresponding handling methods are the same for different event levels, with the only difference being the evidence collection time. Higher time levels correspond to longer evidence collection periods and subsequent evidence retention periods. The specific event level classification method, as well as the relationship between event level, evidence collection period, and evidence retention period, can be determined based on relevant actual circumstances. This application does not impose any specific restrictions.

[0043] Subsequently, an anti-tampering identification layer is added to each data unit in the continuous data segment. This identification layer is overlaid on the original data using digital watermarking technology and includes a data generation time marker, a spatial location marker, and a device identification marker. The continuous data segments with the added anti-tampering identification layer are then integrated into a structured evidence data package according to the time sequence. In this package, video frames and audio clips are strictly aligned with the timeline, and temperature data and motion trajectory information are embedded in the metadata layer in JSON format. This ensures that the time span of the data segment is positively correlated with the event level. For example, in the case of similar abnormal events, the total duration of the evidence data package for a Level 1 event is 40 minutes, for a Level 2 event it is 20 minutes, and for a Level 3 event it is 4 minutes.

[0044] The videos stored in the vehicle's security chip and uploaded to the cloud evidence database may contain the owner's privacy, vehicle location, and personal information of people around them. In order to securely store the evidence data, it is encrypted and anonymized.

[0045] Finally, the generated structured evidence data packets are verified for integrity through the feature fusion layer of the edge computing unit and encrypted block by block using the AES-256 algorithm. The encryption key is dynamically generated by the hardware security module and stored in the vehicle security chip. At the same time, the encrypted data packets are uploaded to the cloud evidence library via the TLS 1.3 protocol and simultaneously stored locally on a tamper-proof encrypted SSD, forming a closed-loop evidence chain for judicial evidence collection standards.

[0046] In one embodiment of this application, before encrypting the evidence data, the method further includes: scanning dynamic image frames in the evidence data based on a preset sensitive information feature library, identifying privacy-sensitive areas containing facial features, vehicle identification features, or biometric behavioral features, and retaining an encrypted backup of the original data to a preset independent security partition; blurring the identified privacy-sensitive areas to generate privacy-desensitized data with masked tags; establishing a mapping relationship between privacy masked tags and event levels, and dynamically adjusting the visualization permissions of the masked areas according to the event levels; encapsulating the privacy-desensitized data and the mapping relationship into a privacy-enhanced evidence unit, and synchronously embedding the mapping relationship as independent metadata during the encryption process; generating a privacy-desensitization operation log, recording the masked area location, operation time, and event level association information, and synchronously storing the privacy-desensitization operation log to a preset monitoring node.

[0047] In one specific embodiment of this application, the following steps are included before encrypting the evidence data: First, dynamic image frames in the evidence data are scanned based on a pre-defined sensitive information feature library to identify privacy-sensitive regions containing facial features, vehicle identification features, or biometric behavioral features. This feature library is generated through deep learning model training and includes face detection algorithms, license plate recognition algorithms, and human pose estimation models. Multi-scale sliding window detection is performed on each frame to extract the facial contours, license plate shapes, and body movement features of the target object. The recognition results are then matched with privacy-sensitive patterns in the feature library. Once a privacy-sensitive region meeting a pre-defined threshold is found, a retention mechanism is immediately triggered. The complete evidence unit containing the original data is encrypted using the AES-256 algorithm and stored in a pre-defined independent secure partition. This partition is separated from the main data stream by hardware isolation technology and is only accessible to supervisory nodes with multi-level authentication. It should be noted that the aforementioned face detection algorithm, license plate recognition algorithm, and human pose estimation model can be implemented using different existing solutions according to the actual needs of the application, and this application does not impose any specific restrictions on them.

[0048] The identified privacy-sensitive areas are then blurred using an adaptive Gaussian blur algorithm to dynamically adjust the mask radius. The mask coverage is calculated based on the target object's movement trajectory, generating privacy-desensitized data with masked tags while retaining an encrypted backup of the original data. The system further establishes a mapping between privacy mask tags and event levels. For example, the masked area for a Level 1 event (fire) is only accessible to judicial institutions through a dedicated decryption channel; the masked area for a Level 2 event (fight) can have its access temporarily lifted during emergency response; and the masked area for a Level 3 event (falling to the ground) is open to medical rescue personnel by default. The privacy-desensitized data and mapping relationship are encapsulated into a privacy-enhanced evidence unit. During the encryption process, the mapping relationship is synchronously embedded as independent metadata. This metadata includes the access control policy corresponding to the event level and the decryption key hash value, ensuring that differentiated desensitization effects are automatically triggered when different user roles access the system.

[0049] Finally, a privacy-de-identified operation log is generated, recording the location of the masked area, the operation time, and the event level association information. This log is synchronously stored to a preset regulatory node through blockchain technology. A hash chain structure is used to ensure that the log is tamper-proof, and a multi-signature mechanism is used to realize the joint audit authority of judicial institutions and vehicle owners, thereby ensuring privacy compliance while meeting the integrity requirements of judicial evidence collection.

[0050] In one embodiment of this application, encrypting evidence data includes: generating an encryption key uniquely bound to the evidence data and storing the encryption key in a protected independent area; dividing the evidence data into multiple data units and performing iterative encryption operations on each data unit based on preset encryption logic; and adding an integrity verification identifier to the encrypted data units to form a tamper-proof evidence package.

[0051] In one specific embodiment of this application, the specific implementation method for encrypting evidence data is as follows: First, a unique encryption key is generated and stored in a protected, independent area. This key is dynamically generated by the Hardware Security Module (HSM) and uses the Elliptic Curve Cryptography (ECC) algorithm to generate a 256-bit key pair. The private key is stored in a dedicated encrypted partition of the vehicle's security chip using physical isolation technology, while the public key is used for subsequent encryption verification. At the same time, the key fingerprint is bound to the metadata of the evidence data through hash chain technology to ensure the indivisibility of the key and the data.

[0052] The evidence data is then divided into multiple data units, the size of which is dynamically adjusted based on a preset encryption logic. For example, data units for Level 1 events (fire) are divided into 10-second time units, Level 2 events (fight) into 5-second time units, and Level 3 events (falling to the ground) into 1-second time units. Each data unit's header embeds an event level identifier and a timestamp sequence number. Iterative encryption operations are performed on each data unit based on the preset encryption logic, using the AES-256 algorithm in Cipher Block Chaining (CBC) mode to encrypt blocks one by one. The encryption strength is adjusted according to the event level: Level 1 events use a triple encryption strategy (encryption-decryption-re-encryption), Level 2 events use a dual encryption strategy (encryption-hash verification), and Level 3 events use a single-layer encryption strategy, namely "encryption-timestamp verification." Dynamic salt and random nonlinear perturbation factors are introduced during the encryption process to ensure that the same data content produces differentiated ciphertext outputs in different encryption cycles.

[0053] An integrity verification identifier is added to the encrypted data unit. A data fingerprint is generated using the SHA-3 hash algorithm, and the fingerprint value is combined with the metadata of the original data unit (including event level, timestamp, and spatial coordinates) to form an anti-tampering identifier. This identifier is appended to the end of the encrypted data unit in the form of a digital signature. The signature key is generated by the HSM module and is decrypted and verified only when authorized by a judicial authority.

[0054] Ultimately, all encrypted data units and integrity verification identifiers are integrated into a tamper-proof evidence package. The evidence package is encapsulated in a hierarchical JSON format, including the main data stream (encrypted data units), metadata layer (event level mapping relationship and privacy mask mark), and verification layer (hash fingerprint and digital signature). The evidence package is uploaded to the cloud evidence library via the TLS 1.3 protocol and simultaneously stored locally on a tamper-proof encrypted SSD, forming a closed-loop evidence chain for judicial evidence collection standards, ensuring the confidentiality, integrity and traceability of data during transmission, storage and access.

[0055] Step S240: Store the encrypted evidence data in a preset storage center for later retrieval.

[0056] In one embodiment of this application, the specific process of storing encrypted evidence data in a preset storage center is as follows: The tamper-proof evidence package is uploaded to a cloud-based evidence repository via the TLS 1.3 protocol, and simultaneously stored locally on a tamper-proof encrypted SSD. The cloud storage employs a distributed redundant architecture to ensure high data availability, while local storage protects the physical medium through hardware isolation technology. The encrypted data packet is encapsulated in a hierarchical JSON format in the cloud. The main data stream contains encrypted data units. The metadata layer integrates event level mapping relationships and privacy mask markers. The verification layer embeds hash fingerprints and digital signatures. During storage, the hardware security module dynamically generates key fingerprints and binds them to the data, ensuring that the key and data are inseparable. The cloud storage center is configured with a multi-level access authentication mechanism, allowing only judicial institutions to access the data through a dedicated decryption channel. Local storage uses a vehicle security chip for access control. All operation records are synchronously generated into a judicial audit log, which includes data storage time, access path, and event level association information, ultimately forming a closed-loop evidence chain for subsequent retrieval and judicial evidence collection.

[0057] In one embodiment of this application, storing encrypted evidence data in a preset storage center includes: mapping the tamper-proof evidence package to an independent storage partition of the preset storage center according to the abnormal event type and event level; constructing a redundant storage topology containing a primary storage node and at least one backup storage node, and synchronously writing the tamper-proof evidence package to the primary storage node and the backup storage node; generating a multi-node association index for the tamper-proof evidence package based on the redundant storage topology, the multi-node association index containing an event level label and a timestamp sequence; performing periodic data verification according to the redundant storage topology, and triggering a data replacement process from the backup storage node to the primary storage node when the verification of the tamper-proof evidence package of the primary storage node is abnormal; and setting access permission rules for the storage partition based on the event level to restrict unauthorized operation commands from accessing the tamper-proof evidence package.

[0058] In one specific embodiment of this application, the encrypted evidence data is stored in a preset storage center. Taking any abnormal event type as an example, the specific implementation method for evidence data storage is as follows: First, based on the event level, the tamper-proof evidence package is mapped to an independent storage partition in a pre-defined storage center. Level 1 events correspond to a dedicated judicial storage area, Level 2 events to an emergency response storage area, and Level 3 events to a medical rescue storage area. Each partition is physically isolated using hardware isolation technology and configured with independent access control policies. Then, a redundant storage topology is constructed, including a primary storage node and at least one backup storage node. A distributed consensus protocol (such as Raft) is used to ensure data synchronization between the primary and backup nodes. When writing the tamper-proof evidence package, hash verification is performed by the feature fusion layer of the edge computing unit before synchronously writing it to both the primary and backup storage nodes. The primary node stores the complete evidence data package, while the backup node only stores the encryption key fingerprint and metadata digest to save storage space.

[0059] Subsequently, a multi-node association index for tamper-proof evidence packages is generated based on this redundant storage topology. The index structure includes event level labels (such as level 1, 2, and 3), timestamp sequences (accurate to milliseconds), master node hash values, and backup node hash values. The index data is distributed across multiple monitoring nodes using blockchain technology to ensure immutability. The system further implements a periodic data verification mechanism, performing hash comparisons between the evidence packages of the master node and backup nodes every 24 hours. When the master node evidence package verification fails, a data replacement process from the backup node to the master node is automatically triggered. During the replacement process, a temporary key is dynamically generated through the Hardware Security Module (HSM) to complete the data decryption and re-encryption operations, ensuring data consistency and key security.

[0060] Finally, access permission rules for storage partitions are set based on event levels. For example, the Level 1 event storage area is only accessible to judicial institutions through multi-level authentication channels, the Level 2 event storage area is open to emergency response departments with limited access periods, and the Level 3 event storage area is allowed by default to be temporarily accessed by medical rescue personnel during the emergency response phase. All access requests must be matched to roles through the permission verification module of the vehicle safety chip. Unauthorized operation instructions will be recorded in the judicial audit log and trigger an alarm mechanism. This ultimately forms a closed-loop management system covering storage, access, verification, and access control to ensure the integrity, availability, and compliance of evidence data in judicial evidence collection scenarios.

[0061] In one embodiment of this application, after obtaining the abnormal event type and its corresponding event level, the method includes triggering a corresponding response strategy based on the event level. For example, an abnormal event type related to fire triggers a fire alarm, directly connects to 119 via V2X, and pushes fire source location data; an abnormal event type related to public security triggers a public security alarm, automatically dials 110, and uploads conflict videos with watermarks; an abnormal event type related to personal safety triggers a linkage alarm, including a public security alarm and a medical alarm, wherein the medical alarm specifically includes calling 120 and broadcasting vital sign information.

[0062] Figure 3 This is a schematic diagram illustrating the complete steps of an emergency response method based on vehicle sentry mode provided in one embodiment of the present invention.

[0063] like Figure 3 As shown, the emergency response method based on vehicle sentry mode proposed in this application includes a technical process for an event-level response mechanism based on environmental data collection and analysis. This process involves steps from virtual scenario construction to actual environmental data processing, ultimately forming a closed-loop evidence chain. The specific process is as follows: First, virtual scene data containing typical abnormal events such as fires, fights, and people falling to the ground are generated using the Unity3D engine to build a multimodal scene library, providing a training foundation for the behavior recognition model. The scene library includes cross-modal data such as video frame motion trajectories, audio voiceprint features, and temperature gradient changes, ensuring that the model has the ability to perceive complex events. Based on this, the behavior recognition model is trained using the scene library data. The model uses a spatiotemporal convolutional neural network to fuse and analyze video, audio, and thermodynamic anomaly features to achieve probabilistic determination of the type of abnormal event.

[0064] Then, in the actual deployment phase, environmental data is collected in real time by vehicle-mounted multimodal sensors such as 4K cameras, microphone arrays, and infrared thermal imaging modules. Spatiotemporal fusion feature vectors, such as target displacement velocity, acoustic spectrum energy, and temperature field gradient changes, are extracted by edge computing units. The feature vectors are then input into a pre-trained behavior recognition model for event probability analysis. The model outputs probability matrices such as fire confidence, fight confidence, and fall confidence. Combined with preset thresholds and duration, it is determined whether an abnormal event is triggered.

[0065] When an event is triggered, a tiered response strategy is executed based on the event level: for a Level 1 fire event, when the temperature gradient change exceeds 15℃ / second, 119 is directly connected and a fire warning sound is played; for a Level 1 brawl event, when the brawl confidence level exceeds 85%, 110 is automatically dialed and a siren is played; for a Level 1 fall event, when a person falls to the ground for more than 3 minutes or does not move after falling, 120 is called and the vehicle's coordinates and status are announced.

[0066] During this process, for the same type of abnormal event, a preset time window parameter is matched according to the event level (e.g., for a level 1 event, data from 30 minutes before the event to 10 minutes after the event is retained). Continuous data segments within the target time interval are extracted from the environmental data stream and overlaid with an anti-tampering identification layer (including timestamps, GPS coordinates, and device ID watermarks). The extracted evidence data is encrypted block by block using the AES-256 algorithm. The key is dynamically generated by the Hardware Security Module (HSM) and synchronously written to the cloud evidence repository and the local anti-tampering SSD via the TLS 1.3 protocol. At the same time, a structured evidence data package is generated, including the main data stream (encrypted data units), the metadata layer (event level mapping relationship and privacy mask marking), and the verification layer (hash fingerprint and digital signature).

[0067] Ultimately, by constructing a redundant topology of primary storage nodes and backup storage nodes, and implementing periodic data verification and access control, the integrity, availability, and compliance of evidence data in judicial evidence collection scenarios are ensured, forming a closed-loop management system covering the entire process of event identification, response, evidence generation, and storage.

[0068] It is also important to emphasize that the emergency response method based on vehicle sentry mode proposed in this application achieves graded responses to events such as fires, fights, and falls by fusing video, audio, and thermodynamic features through the fusion analysis of onboard sensors and AI models. This significantly improves identification accuracy and reduces false alarms. After an event is triggered, environmental data is dynamically captured and overlaid with tamper-proof identifiers. This data is then encrypted using AES-256 and bound to a hardware security module to generate a structured evidence package, which is simultaneously written to the cloud and redundant local storage to ensure evidence integrity and judicial compliance. Sensitive areas are dynamically blurred, access permissions are set based on event levels, and de-identified operation logs are recorded using blockchain technology to ensure privacy compliance. In summary, this method achieves intelligent, compliant, and closed-loop management of vehicle security protection through AI-driven proactive defense and judicial-grade management.

[0069] Figure 4 This is a block diagram illustrating an emergency response device based on vehicle sentry mode, as shown in an exemplary embodiment of this application. This device can be applied to, for example... Figure 1 The implementation environment shown is illustrated. This device can also be applied to other exemplary implementation environments and specifically configured in other devices. This embodiment does not limit the implementation environment to which the device is applicable.

[0070] like Figure 4 As shown, the exemplary emergency response device based on vehicle sentry mode includes: a multimodal sensing module 410, an edge computing module 420, a response execution module 430, an evidence chain management module 40, and a hierarchical storage module 450.

[0071] The system includes a multimodal sensing module 410 for collecting environmental data around the vehicle; an edge computing module 420 for generating a spatiotemporal fusion feature vector of the environmental data and determining the abnormal event type and event level based on the spatiotemporal fusion feature vector; a behavior recognition model trained on scenarios defined by the environmental data; a response execution module 430 for triggering a hierarchical response protocol based on the abnormal event type and the event level; an evidence chain management module 440 for capturing environmental data within a time window matched to the event level and encrypting the evidence data based on a preset encryption algorithm; and a hierarchical storage module 450 for storing the corresponding encrypted evidence data according to the event level.

[0072] In one specific embodiment of this application, the evidence chain management module includes: an anti-tampering identification unit, a privacy desensitization unit, a mask processing subunit, and an original data backup subunit.

[0073] The system includes: an anti-tampering identification unit for overlaying a watermark layer containing time markers, spatial location markers, and device identification markers onto the captured environmental data; a privacy desensitization unit for scanning privacy-sensitive areas containing facial features, vehicle identification features, or biological behavioral features in dynamic image frames based on a feature library; a masking processing subunit for blurring the information in the privacy-sensitive areas and generating corresponding privacy-desensitized data; and an original data backup subunit for encrypting and storing the undesensitized data in the privacy-sensitive areas in a preset independent secure partition.

[0074] It should be noted that the emergency response device based on vehicle sentry mode provided in the above embodiments and the emergency response method based on vehicle sentry mode provided in the above embodiments belong to the same concept. The specific operation methods of each module and unit have been described in detail in the method embodiments and will not be repeated here. In practical applications, the emergency response device based on vehicle sentry mode provided in the above embodiments can be assigned to different functional modules as needed, that is, the internal structure of the device can be divided into different functional modules to complete all or part of the functions described above. This is not a limitation here.

[0075] Embodiments of this application also provide an electronic device, including: one or more processors; and a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the emergency response method based on vehicle sentry mode provided in the above embodiments.

[0076] Figure 5 A schematic diagram of a computer system suitable for implementing the embodiments of this application is shown. It should be noted that... Figure 5The computer system 500 of the electronic device shown is merely an example and should not impose any limitation on the functionality and scope of use of the embodiments of this application.

[0077] like Figure 5 As shown, the computer system 500 includes a Central Processing Unit (CPU) 501, which can perform various appropriate actions and processes based on programs stored in Read-Only Memory (ROM) 502 or programs loaded from storage portion 508 into Random Access Memory (RAM) 503, such as performing the methods described in the above embodiments. The RAM 503 also stores various programs and data required for system operation. The CPU 501, ROM 502, and RAM 503 are interconnected via a bus 504. An Input / Output (I / O) interface 505 is also connected to the bus 504.

[0078] The following components are connected to I / O interface 505: an input section 506 including a keyboard, mouse, etc.; an output section 507 including a cathode ray tube (CRT), liquid crystal display (LCD), etc., and speakers, etc.; a storage section 508 including a hard disk, etc.; and a communication section 509 including a network interface card such as a LAN (Local Area Network) card, modem, etc. The communication section 509 performs communication processing via a network such as the Internet. A drive 510 is also connected to I / O interface 505 as needed. Removable media 511, such as a disk, optical disk, magneto-optical disk, semiconductor memory, etc., are installed on drive 510 as needed so that computer programs read from them can be installed into storage section 508 as needed.

[0079] Specifically, according to embodiments of this application, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this application include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication section 509, and / or installed from removable medium 511. When the computer program is executed by central processing unit (CPU) 501, it performs various functions defined in the system of this application.

[0080] It should be noted that the computer-readable medium shown in the embodiments of this application can be a computer-readable signal medium or a computer-readable storage medium, or any combination of the two. A computer-readable storage medium can be, for example, 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), flash memory, optical fiber, portable compact disc read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this application, a computer-readable signal medium may include a data signal propagated in baseband or as part of a carrier wave, carrying a computer-readable computer program. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. Computer-readable signal media can also be any computer-readable medium other than computer-readable storage media, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The computer program contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to wireless, wired, etc., or any suitable combination thereof.

[0081] 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 application. Each block in a flowchart or block diagram may represent a module, segment, or portion of code, which contains 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 a block diagram or flowchart, and combinations of blocks in a block diagram or flowchart, 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.

[0082] The units involved in the embodiments of this application can be implemented in software or hardware, and the described units can also be located in a processor. The names of these units do not necessarily limit the unit itself under certain circumstances.

[0083] Another aspect of this application provides a computer-readable storage medium storing a computer program thereon, which, when executed by a computer's processor, causes the computer to perform the emergency response method based on vehicle sentry mode as described above. This computer-readable storage medium may be included in the electronic device described in the above embodiments, or it may exist independently and not incorporated into the electronic device.

[0084] Another aspect of this application provides a computer program product or computer program including computer instructions stored in a computer-readable storage medium. A processor of a computer device reads the computer instructions from the computer-readable storage medium and executes the computer instructions, causing the computer device to perform the emergency response method based on vehicle sentry mode provided in the various embodiments described above.

[0085] The above embodiments are merely illustrative of the principles and effects of the present invention and are not intended to limit the invention. Any person skilled in the art can modify or alter the above embodiments without departing from the spirit and scope of the present invention. Therefore, all equivalent modifications or alterations made by those skilled in the art without departing from the spirit and technical concept disclosed in the present invention should still be covered by the claims of the present invention.

Claims

1. An emergency response method based on vehicle sentry mode, characterized in that, The method includes: Collect environmental data of the vehicle's location, including at least video data, audio data, and thermal radiation data; The environmental data is input into a preset behavior recognition model to obtain the abnormal event type and the corresponding event level. The behavior recognition model is trained by dividing the environmental data into scenarios. Based on the abnormal event type and the event level, extract the corresponding evidence data from the environmental data and encrypt the evidence data. The encrypted evidence data is stored in a pre-defined storage center.

2. The emergency response method based on vehicle sentry mode according to claim 1, characterized in that, Obtain the exception event type and its corresponding event level, including: Multi-dimensional feature extraction is performed on the collected environmental data to generate a spatiotemporal fusion feature vector; The spatiotemporal fusion feature vector is input into a preset behavior recognition model to perform event probability analysis, and the confidence parameters corresponding to the abnormal event types are obtained. When the confidence parameter of an abnormal event type exceeds a preset threshold and the duration exceeds a preset duration threshold, it is determined that the abnormal event type has been triggered. Based on the type of abnormal event, the duration of continuously triggered abnormal events is monitored, and the event level is determined according to the duration.

3. The emergency response method based on vehicle sentry mode according to claim 1, characterized in that, Extract relevant evidence data from environmental data, including: Based on the abnormal event type and the event level, a preset time window parameter is matched to determine the start and end time points of the data segment to be extracted; Based on the start time point and the end time point, extract continuous data segments within the target time interval from the continuously collected environmental data stream; An anti-tampering identification layer is added to each data unit in the continuous data segment. The anti-tampering identification layer includes a watermark superimposed with a data generation time marker, a spatial location marker, and a device identification marker. Continuous data segments with added tamper-proof identification layers are integrated into a structured evidence data package according to time sequence, where the time span of the data segments is positively correlated with the event level.

4. The emergency response method based on vehicle sentry mode according to claim 1, characterized in that, Encrypting the evidence data includes: Generate an encryption key uniquely bound to the evidence data, and store the encryption key in a protected, independent area. The evidence data is divided into multiple data units, and each data unit is subjected to iterative encryption operations based on preset encryption logic. Add an integrity verification identifier to the encrypted data unit to form a tamper-proof evidence package.

5. The emergency response method based on vehicle sentry mode according to claim 4, characterized in that, The encrypted evidence data is stored in a pre-designated storage center, including: The tamper-proof evidence package is mapped to an independent storage partition in a preset storage center according to the type and level of the abnormal event. Construct a redundant storage topology structure that includes a primary storage node and at least one backup storage node, and synchronously write the tamper-proof evidence package into the primary storage node and the backup storage node; Based on the redundant storage topology, a multi-node association index for the tamper-proof evidence package is generated. The multi-node association index includes event level tags and timestamp sequences. Periodic data verification is performed according to the redundant storage topology. When the verification of the tamper-proof evidence package of the primary storage node is abnormal, a data replacement process from the backup storage node to the primary storage node is triggered. Access rules for the storage partition are set based on the event level to restrict unauthorized operation commands from accessing the tamper-proof evidence package.

6. The emergency response method based on vehicle sentry mode according to any one of claims 1-5, characterized in that, Before encrypting the evidence data, the process also includes: Based on a pre-defined sensitive information feature database, the system scans dynamic image frames in the evidence data to identify privacy-sensitive areas containing facial features, vehicle identification features, or biometric behavioral features, and retains encrypted backups of the original data to a pre-defined independent security partition. The identified privacy-sensitive areas are blurred to generate privacy-de-identified data with masked tags; Establish a mapping relationship between privacy mask markers and event levels, and dynamically adjust the visualization permissions of the mask area according to the event level; Privacy-de-identified data and mapping relationships are encapsulated into privacy-enhanced evidence units, and the mapping relationships are synchronously embedded as independent metadata during the encryption process. Generate a privacy desensitization operation log, recording the mask area location, operation time, and event level association information, and synchronously store the privacy desensitization operation log to a preset monitoring node.

7. An emergency response device based on vehicle sentry mode, characterized in that, The device includes: A multimodal sensing module is used to collect data on the vehicle's surrounding environment. The edge computing module is used to generate a spatiotemporal fusion feature vector of environmental data, and to determine the abnormal event type and event level based on the spatiotemporal fusion feature vector. The behavior recognition model is trained by dividing the environmental data into scenarios. A response execution module is used to trigger a hierarchical response protocol based on the abnormal event type and the event level. The evidence chain management module is used to capture environmental data according to the time window matched by the event level, and to encrypt the evidence data based on a preset encryption algorithm. A tiered storage module is used to retain corresponding encrypted evidence data according to the event level.

8. The emergency response device based on vehicle sentry mode according to claim 7, characterized in that, The evidence chain management module includes: The tamper-proof identification unit is used to overlay a watermark layer containing time markers, spatial location markers, and device identification markers onto the captured environmental data; The privacy desensitization unit is used to scan privacy-sensitive areas containing facial features, vehicle identification features, or biometric behavioral features in dynamic image frames based on a feature library. The masking subunit is used to blur the information in the privacy-sensitive area and generate corresponding privacy-de-identified data; The original data backup subunit is used to encrypt and store the un-de-identified data of the privacy-sensitive area to a preset independent security partition.

9. An electronic device, characterized in that, The electronic device includes: One or more processors; A storage device for storing one or more programs that, when executed by one or more processors, cause the electronic device to implement the emergency response method based on vehicle sentry mode as described in any one of claims 1 to 6.

10. A computer-readable storage medium, characterized in that, It stores a computer program that, when executed by the computer's processor, causes the computer to perform the emergency response method based on vehicle sentry mode as described in any one of claims 1 to 6.

Images