Intelligent monitoring method and system based on large model

Abstract

The embodiment of the application discloses a kind of intelligent monitoring method and system based on large model;Method includes: receiving the data to be processed sent by intelligent terminal;Based on the system prompt word template set in advance, input data is generated according to preprocessed sensor data and structured prompt word;Input data is input into large model to analyze and reason, and structured output data is generated;Based on structured output data, alarm information is sent to user and / or external system.The application has the following advantages: users directly describe monitoring requirements through natural language, the system processes it structurally, makes the explicitness of monitoring object, condition, threshold value, etc., so that large model and system logic can accurately understand user intent, improve the degree of freedom and accuracy of monitoring task configuration;Fusion monitoring is realized based on multiple sensors, and various scenarios are met;Alarm information is sent to various external systems in various ways, which can realize external automatic linkage and expand the application scenarios of monitoring results.

Description

Technical Field

[0001] This invention relates to the field of intelligent monitoring technology, specifically to an intelligent monitoring method and system based on a large model. Background Technology

[0002] With the development of mobile terminals and IoT technology, video surveillance systems based on camera equipment have been widely used in homes, offices, industries, public places, and other fields. Traditional surveillance systems typically use fixed-function monitoring algorithms, such as personnel intrusion detection, traffic flow statistics, and facial recognition. Users can only choose from a limited number of preset functions, making it difficult to flexibly configure personalized monitoring needs.

[0003] The existing technology has at least the following problems:

[0004] 1. Insufficient flexibility in monitoring tasks

[0005] In most surveillance systems, the monitoring scenarios and objectives are pre-defined by the system developers, and users cannot freely describe their monitoring needs using natural language. For example, requirements such as "remind me when the dog in the video gets on the sofa" or "remind me when the indoor temperature exceeds 28°C for 5 minutes" are often not achievable through simple configuration in existing systems.

[0006] 2. Insufficient multi-sensor fusion capability

[0007] Traditional monitoring systems primarily rely on video images as their data source, making limited use of data from other sensors such as temperature, vibration, and equipment posture. This makes it difficult to monitor comprehensive scenarios such as abnormal ambient temperature, equipment movement, and abnormal vibration.

[0008] 3. Lack of a mechanism for standardizing prompt words

[0009] User-input natural language has problems such as vague expression and incomplete semantics. Existing systems generally lack mechanisms to standardize and structure natural language prompts, making it difficult for subsequent algorithms and models to accurately understand the user's monitoring intent.

[0010] 4. Unintelligent authorization management and insufficient privacy protection

[0011] Existing systems often require users to manually authorize sensitive sensors such as cameras and microphones one by one, lacking a mechanism to automatically recommend necessary sensor permissions based on the monitoring scenario. This increases the complexity of user operations and may lead to the excessive collection of user privacy data.

[0012] 5. The data acquisition and transmission strategy is extensive.

[0013] Most systems use fixed sampling frequencies and data upload strategies, lacking differentiated processing for scenarios with different levels of urgency, making it difficult to achieve a good balance between real-time monitoring and terminal power consumption and network bandwidth consumption.

[0014] 6. The alarm methods are limited and lack system linkage capabilities.

[0015] Existing monitoring systems often limit their alarm methods to local alerts or simple notification pushes, lacking a unified multi-channel alarm and priority management mechanism. Furthermore, they typically do not support sending alarm data to external systems via APIs, thus limiting their ability to integrate with other systems (such as smart home systems, security systems, and enterprise business systems). Summary of the Invention

[0016] The purpose of this invention is to provide an intelligent monitoring method and system based on a large model, which can simultaneously achieve functions such as natural language task configuration, multi-sensor data fusion, prompt word standardization processing, intelligent authorization management, adaptive acquisition and transmission, multi-channel reminders and API linkage.

[0017] To achieve the above objectives, in a first aspect, embodiments of the present invention provide an intelligent monitoring method based on a large model, comprising:

[0018] Receive data to be processed sent by a smart terminal; the data to be processed includes preprocessed sensor data and structured prompts.

[0019] Based on a pre-set system prompt word template, input data is generated according to pre-processed sensor data and structured prompt words;

[0020] The input data is fed into a large model for parsing and reasoning to generate structured output data;

[0021] Alarm information is sent to users and / or external systems based on the structured output data.

[0022] As one specific implementation of this application, the process of generating the preprocessed sensor data is as follows:

[0023] The application on the smart terminal receives monitoring request prompts and configuration data input by the user; the configuration data is the user's configuration of the monitoring scenario type and urgency level.

[0024] The monitoring request prompts are processed to obtain structured prompts; the structured prompts include the monitoring object, monitoring conditions, threshold parameters, duration, and urgency of the scenario.

[0025] Receive user authorization for the use of the sensor and obtain authorization information;

[0026] Sensor data is collected based on a preset sampling strategy, and the sensor data is preprocessed.

[0027] As one specific implementation of this application, sensor data is collected based on a preset sampling strategy, specifically as follows:

[0028] Based on the structured prompts and the authorization information, a sampling strategy for the corresponding sensor is determined; the sampling strategy includes sampling frequency and / or data upload strategy.

[0029] Sensor data from at least one sensor is collected based on the sampling strategy described above; the sensor includes a camera, a microphone, a temperature sensor, a humidity sensor, and a gyroscope sensor.

[0030] As a specific implementation of this application, alarm information is sent to the user and / or external system based on the structured output data, specifically as follows:

[0031] The structured output data is parsed and judged. When the alarm triggering conditions are met, alarm information is sent to the user and / or external system through an alarm method according to the pre-configured multi-channel reminder strategy and reminder priority. The alarm method includes at least one of SMS, email, in-app push notification, smart terminal vibration, smart terminal speaker prompt sound and API interface.

[0032] Secondly, embodiments of the present invention also provide an intelligent monitoring system based on a large model, including an intelligent terminal and a server; the intelligent terminal is used to: collect data to be processed and send the data to be processed to the server; the data to be processed includes preprocessed sensor data and structured prompt words;

[0033] The server is used for:

[0034] Receive the data to be processed;

[0035] Based on a pre-set system prompt word template, input data is generated according to pre-processed sensor data and structured prompt words;

[0036] The input data is fed into a large model for parsing and reasoning to generate structured output data;

[0037] Alarm information is sent to users and / or external systems based on the structured output data.

[0038] As one specific implementation of this application, the smart terminal includes:

[0039] The user interface is used to receive monitoring request prompts and configuration data input by the user; the configuration data is the user's configuration of the monitoring scenario type and urgency level.

[0040] The standardization processing unit is used to standardize the monitoring requirement prompts to obtain structured prompts; the structured prompts include the monitoring object, monitoring conditions, threshold parameters, duration, and scene urgency.

[0041] The authorization unit is used to receive user authorization for the use of the sensor and obtain authorization information.

[0042] The data acquisition unit is used to acquire sensor data based on a preset sampling strategy;

[0043] The preprocessing unit is used to preprocess the sensor data, including compression, sampling, filtering, format conversion, and time alignment.

[0044] The data transmission unit is used to send the preprocessed sensor data and structured prompts to the server.

[0045] Furthermore, the data acquisition unit is specifically used for:

[0046] Based on the structured prompts and the authorization information, a sampling strategy for the corresponding sensor is determined; the sampling strategy includes sampling frequency and / or data upload strategy.

[0047] Sensor data from at least one sensor is collected based on the sampling strategy described above; the sensor includes a camera, a microphone, a temperature sensor, a humidity sensor, and a gyroscope sensor.

[0048] As one specific implementation of this application, the server includes:

[0049] A data receiving unit is used to receive data to be processed from multiple smart terminals and to classify and store the data to be processed.

[0050] The data processing unit is used to generate input data based on pre-processed sensor data and structured prompts, according to a pre-set system prompt template.

[0051] The large model invocation unit is used to invoke the large model to parse and reason about the input data to obtain structured output data; the structured output data includes at least one of the following: indication information on whether the monitoring conditions are met, event type, confidence level, and cause explanation.

[0052] As one specific implementation of this application, the server further includes an alarm unit, used for:

[0053] The structured output data is parsed and judged. When the alarm triggering conditions are met, alarm information is sent to the user and / or external system through an alarm method according to the pre-configured multi-channel reminder strategy and reminder priority. The alarm method includes at least one of SMS, email, in-app push notification, smart terminal vibration, smart terminal speaker prompt sound and API interface.

[0054] The advantages of implementing the intelligent monitoring scheme based on a large model provided in the embodiments of the present invention are as follows:

[0055] 1. Natural Language Task Configuration and Prompt Word Standardization: Users can directly describe their monitoring needs using natural language. The system performs semantic parsing and structured processing on the prompt words, making the monitoring objects, conditions, thresholds, etc. explicit. This allows both the large model and the system logic to accurately understand the user's intent, improving the freedom and accuracy of monitoring task configuration.

[0056] 2. Multi-sensor data fusion monitoring

[0057] By utilizing various sensors such as cameras, microphones, temperature sensors, accelerometers, and gyroscopes, it can comprehensively analyze multi-dimensional data such as images, audio, ambient temperature, and device motion status, meeting the needs of various scenarios such as pet behavior monitoring, temperature anomaly monitoring, and device movement monitoring.

[0058] 3. Intelligent authorization management and privacy protection

[0059] The system automatically recommends the types of sensors that need to be authorized based on the type of monitoring scenario. Users only need to confirm the necessary authorizations, simplifying the configuration process, avoiding the over-activation of irrelevant sensors, and enhancing user privacy protection.

[0060] 4. Adaptive Sampling and Transmission Strategy

[0061] Select high-frequency or low-frequency sampling and corresponding data upload strategies based on the urgency of the monitoring scenario. This ensures real-time monitoring of high-risk events while reducing the consumption of power and network resources in non-urgent scenarios.

[0062] 5. Structured Large Model Input / Output Design

[0063] By constraining the input and output formats of large models using system prompt word templates, the structured data output by large models can be automatically parsed and used for decision-making by the server, thereby improving system stability and scalability.

[0064] 6. Multi-channel, multi-priority alerts and API integration

[0065] Alarms can reach users through various means such as SMS, email, push notifications, vibration, and speakerphone, and can be sent to external systems in the form of data via API interface, realizing automatic linkage with air conditioning control, security systems, enterprise business systems, etc., thus expanding the application scenarios of monitoring results. Attached Figure Description

[0066] To more clearly illustrate the specific embodiments of the present invention or the technical solutions in the prior art, the accompanying drawings used in the description of the specific embodiments or the prior art will be briefly introduced below.

[0067] Figure 1 This is a flowchart of the intelligent monitoring method based on a large model provided in an embodiment of the present invention;

[0068] Figure 2 This is a structural diagram of the intelligent monitoring system based on a large model provided in an embodiment of the present invention;

[0069] Figure 3 yes Figure 2 The diagram shows another architecture of the server. Detailed Implementation

[0070] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0071] It should be understood that, when used in this specification and the appended claims, the terms "comprising" and "including" indicate the presence of the described features, integrals, steps, operations, elements and / or components, but do not exclude the presence or addition of one or more other features, integrals, steps, operations, elements, components and / or collections thereof.

[0072] To address the technical problems mentioned in the background section, embodiments of the present invention provide a large-scale intelligent monitoring scheme based on natural language task configuration and multi-sensor data. For example... Figure 1 As shown, the intelligent monitoring method based on a large model provided in this embodiment of the invention includes the following process:

[0073] S1, in the application of the smart terminal, the smart terminal receives the monitoring request prompts and configuration data input by the user.

[0074] In practice, an application is installed on the smart terminal, providing a user interface. On this interface, users can input monitoring request prompts in natural language and configuration data. This configuration data includes, but is not limited to, the monitored object, monitoring conditions, selectable threshold parameters, duration, and the urgency level of the scenario.

[0075] S2, perform prompt word standardization processing on the monitoring requirement prompt words to obtain structured prompt words.

[0076] In practice, the user's original natural language prompts are processed through word segmentation, semantic parsing, and entity recognition to extract information such as the monitored object, monitoring conditions, threshold parameters, duration, and urgency of the scenario, generating structured prompts. These structured prompts can be represented in key-value pair format, such as a monitored object field, a condition field, and a trigger threshold field.

[0077] S3: Receive user authorization for the use of the sensor, obtain authorization information, collect sensor data based on a preset sampling strategy, and preprocess the sensor data.

[0078] In practice, based on the preset mapping relationship between monitoring scene types and sensor types, after the user inputs monitoring requirements and scene type, the system automatically deduces and recommends the sensors that need to be enabled, such as cameras, temperature sensors, accelerometers, gyroscopes, etc., and records the user's authorization results.

[0079] Subsequently, based on the structured prompts and the sensor authorization information, a corresponding sensor sampling strategy is determined, wherein the sampling strategy includes at least one of sampling frequency and data upload strategy.

[0080] Next, the smart terminal collects sensor data from the at least one hardware sensor according to the sampling strategy. The hardware sensor includes at least one of a camera, microphone, temperature sensor, accelerometer, and gyroscope sensor. The sensor data includes, but is not limited to, image data, audio features, temperature data sequences, acceleration data sequences, and gyroscope data sequences.

[0081] Finally, the collected sensor data undergoes preprocessing such as compression, sampling, filtering, format conversion, and time alignment to reduce data volume and standardize data format. For example, image resolution is compressed, video frames are extracted, audio is downsampled, and acceleration and gyroscope data are aggregated by time window.

[0082] S4 sends the preprocessed sensor data and structured prompts to the server.

[0083] In practice, based on the data upload strategy in the sampling strategy and the current network status, the preprocessed sensor data, together with the structured prompts for the corresponding task, is encapsulated into a data packet and sent to the server-side monitoring backend. The data can be cached locally when the network is unstable and resent after the network is restored.

[0084] It should be noted that, in some embodiments, the smart terminal can perform local threshold pre-filtering on the accelerometer and gyroscope sensors. When a change exceeding a preset threshold is detected, it triggers an increase in the camera data acquisition frequency and immediately uploads the monitoring data to enhance the response capability to device movement or vibration events.

[0085] S5: The server generates input data based on the pre-set system prompt word template, pre-processed sensor data, and structured prompt words.

[0086] In practice, the server constructs a data structure suitable for large model inputs based on pre-configured system prompt word templates, combined with structured prompt words and pre-processed sensor data.

[0087] Among them, the system prompt word template can clearly define the role, task description, input field description and output format requirements of the large model, so as to guide the large model to stably output structured results.

[0088] S6, the server inputs the input data into the large model for parsing and reasoning, and generates structured output data.

[0089] In practice, the server calls the large model via a network interface, inputs the data into the large model for parsing and inference, and outputs structured data. This structured output includes, but is not limited to, indications of whether monitoring conditions are met, event types, confidence levels, and explanations of causes.

[0090] For example, the structured data output by large models should preferably be in JSON format, including but not limited to the following fields:

[0091] trigger: A boolean value used to indicate whether the monitoring conditions are currently met;

[0092] event_type: A string used to represent the type of event identified, such as "pet_on_sofa", "temperature_high", "device_moved", etc.

[0093] confidence: A numerical value used to represent the confidence level of the judgment result;

[0094] reason: A string used to explain the basis for the judgment;

[0095] recommended_action: An optional string used to suggest the recommended follow-up action, such as sending push notifications or sending API alerts.

[0096] S7, the server sends alarm information to the user and / or external system based on the structured output data.

[0097] In specific implementation, the structured output data is parsed and judged. When it is determined that the alarm triggering condition is met, alarm information is sent to the user and / or external system through an alarm method according to the pre-configured multi-channel reminder strategy and reminder priority. The alarm method includes at least one of SMS, email, in-app push notification, smart terminal vibration, smart terminal speaker prompt sound, and API interface.

[0098] Combining the aforementioned steps S1-S7, this invention provides several monitoring scenarios:

[0099] Scenario 1: Monitoring scene of pets on sofa

[0100] The user inputs the prompt phrase, "Remind me when the puppy in the picture gets on the sofa," selects the scenario type as home pet monitoring, and sets the urgency level to "normal." The smart device is recommended to enable camera authorization. The system generates structured prompts and sets an appropriate image acquisition frequency strategy. The large-scale model determines whether the puppy is in the sofa area based on the image data and sends an alert via in-app push notifications and device vibration when the conditions are met.

[0101] Scenario 2: Temperature Exceedance Monitoring and Air Conditioning System Linkage

[0102] The user inputs the prompt: "Remind me when the room temperature exceeds 28 degrees Celsius and notify the air conditioning system to cool down." The user selects environmental monitoring as the scenario type and high urgency as the priority. The terminal activates the temperature sensor, collecting temperature data at high frequency. When the large model outputs an event type of "temperature_high" and the trigger is true, the system sends an alarm to the user via email and SMS, and sends the alarm data to the air conditioning control system via a pre-configured API to trigger automatic cooling.

[0103] Scenario 3: Monitoring of abnormal equipment movement / vibration

[0104] The user inputs the prompt phrase, "Alert me when my phone is moved or violently vibrates," activating the accelerometer and gyroscope sensors. The device performs threshold pre-filtering on motion data locally, uploading relevant data when abnormal changes are detected. After analysis by a large model, the event is identified as "device_moved," triggering high-priority SMS and push notification alerts.

[0105] As can be seen from the above description, the advantages of implementing the intelligent monitoring scheme based on a large model provided in the embodiments of the present invention are as follows:

[0106] 1. Natural Language Task Configuration and Prompt Word Standardization: Users can directly describe their monitoring needs using natural language. The system performs semantic parsing and structured processing on the prompt words, making the monitoring objects, conditions, thresholds, etc. explicit. This allows both the large model and the system logic to accurately understand the user's intent, improving the freedom and accuracy of monitoring task configuration.

[0107] 2. Multi-sensor data fusion monitoring

[0108] By utilizing various sensors such as cameras, microphones, temperature sensors, accelerometers, and gyroscopes, it can comprehensively analyze multi-dimensional data such as images, audio, ambient temperature, and device motion status, meeting the needs of various scenarios such as pet behavior monitoring, temperature anomaly monitoring, and device movement monitoring.

[0109] 3. Intelligent authorization management and privacy protection

[0110] The system automatically recommends the types of sensors that need to be authorized based on the type of monitoring scenario. Users only need to confirm the necessary authorizations, simplifying the configuration process, avoiding the over-activation of irrelevant sensors, and enhancing user privacy protection.

[0111] 4. Adaptive Sampling and Transmission Strategy

[0112] Select high-frequency or low-frequency sampling and corresponding data upload strategies based on the urgency of the monitoring scenario. This ensures real-time monitoring of high-risk events while reducing the consumption of power and network resources in non-urgent scenarios.

[0113] 5. Structured Large Model Input / Output Design

[0114] By constraining the input and output formats of large models using system prompt word templates, the structured data output by large models can be automatically parsed and used for decision-making by the server, thereby improving system stability and scalability.

[0115] 6. Multi-channel, multi-priority alerts and API integration

[0116] Alarms can reach users through various means such as SMS, email, push notifications, vibration, and speakerphone, and can be sent to external systems in the form of data via API interface, realizing automatic linkage with air conditioning control, security systems, enterprise business systems, etc., thus expanding the application scenarios of monitoring results.

[0117] Based on the same inventive concept, embodiments of the present invention also provide, as follows: Figure 2The large-scale intelligent monitoring system shown includes an intelligent terminal and a server; the intelligent terminal is used to: collect data to be processed and send the data to be processed to the server; the data to be processed includes preprocessed sensor data and structured prompts;

[0118] The server is used for:

[0119] Receive the data to be processed;

[0120] Based on a pre-set system prompt word template, input data is generated according to pre-processed sensor data and structured prompt words;

[0121] The input data is fed into a large model for parsing and reasoning to generate structured output data;

[0122] Alarm information is sent to users and / or external systems based on the structured output data.

[0123] The smart terminal includes:

[0124] The user interface is used to receive monitoring request prompts and configuration data input by the user; the configuration data is the user's configuration of the monitoring scenario type and urgency level.

[0125] The standardization processing unit is used to standardize the monitoring requirement prompts to obtain structured prompts; the structured prompts include the monitoring object, monitoring conditions, threshold parameters, duration, and scene urgency.

[0126] The authorization unit is used to receive user authorization for the use of the sensor and obtain authorization information.

[0127] The data acquisition unit is used to acquire sensor data based on a preset sampling strategy;

[0128] The preprocessing unit is used to preprocess the sensor data, including compression, sampling, filtering, format conversion, and time alignment.

[0129] The data transmission unit is used to send the preprocessed sensor data and structured prompts to the server.

[0130] Furthermore, the data acquisition unit is specifically used for:

[0131] Based on the structured prompts and the authorization information, a sampling strategy for the corresponding sensor is determined; the sampling strategy includes sampling frequency and / or data upload strategy.

[0132] Sensor data from at least one sensor is collected based on the sampling strategy described above; the sensor includes a camera, a microphone, a temperature sensor, a humidity sensor, and a gyroscope sensor.

[0133] In this embodiment, the server includes:

[0134] A data receiving unit is used to receive data to be processed from multiple smart terminals and to classify and store the data to be processed.

[0135] The data processing unit is used to generate input data based on pre-processed sensor data and structured prompts, according to a pre-set system prompt template.

[0136] The large model invocation unit is used to invoke the large model to parse and reason about the input data to obtain structured output data; the structured output data includes at least one of the following: indication information on whether the monitoring conditions are met, event type, confidence level, and cause explanation;

[0137] Alarm unit, used for:

[0138] The structured output data is parsed and judged. When the alarm triggering conditions are met, alarm information is sent to the user and / or external system through an alarm method according to the pre-configured multi-channel reminder strategy and reminder priority. The alarm method includes at least one of SMS, email, in-app push notification, smart terminal vibration, smart terminal speaker prompt sound and API interface.

[0139] It should be noted that the specific workflow of this embodiment is described in the foregoing method embodiment section, and will not be repeated here.

[0140] Furthermore, another embodiment of the present invention provides another type of server. For example... Figure 3 As shown, the product similarity analysis device based on attribute distance may include: one or more processors 101, one or more input devices 102, one or more output devices 103, and a memory 104. The processors 101, input devices 102, output devices 103, and memory 104 are interconnected via a bus 105. The memory 104 stores a computer program, which includes program instructions. The processor 101 is configured to invoke the program instructions to execute the method described in the above-described method embodiment.

[0141] It should be understood that, in this embodiment of the invention, the processor 101 may be a central processing unit (CPU), but it may also be other general-purpose processors, digital signal processors (DSPs), application-specific integrated circuits (ASICs), field-programmable gate arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor may be a microprocessor or any conventional processor.

[0142] Input device 102 may include a keyboard, etc., and output device 103 may include a display (LCD, etc.), a speaker, etc.

[0143] The memory 104 may include read-only memory and random access memory, and provides instructions and data to the processor 101. A portion of the memory 104 may also include non-volatile random access memory. For example, the memory 104 may also store device type information.

[0144] In specific implementations, the processor 101, input device 102, and output device 103 described in the embodiments of the present invention can execute the implementation methods described in the embodiments of the intelligent monitoring method based on large models provided in the embodiments of the present invention, which will not be repeated here.

[0145] Accordingly, embodiments of the present invention provide a computer-readable storage medium storing a computer program, the computer program including program instructions, which, when executed by a processor, implement the above-described intelligent monitoring method based on a large model.

[0146] The computer-readable storage medium can be an internal storage unit of the system described in any of the foregoing embodiments, such as the system's hard disk or memory. The computer-readable storage medium can also be an external storage device of the system, such as a plug-in hard disk, Smart Media Card (SMC), Secure Digital (SD) card, or Flash Card. Furthermore, the computer-readable storage medium can include both internal storage units and external storage devices. The computer-readable storage medium is used to store the computer program and other programs and data required by the system. The computer-readable storage medium can also be used to temporarily store data that has been output or will be output.

[0147] Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementations should not be considered beyond the scope of this invention.

[0148] In the several embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative. For instance, the division of units is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple units or components may be combined or integrated into another system, or some features may be ignored or not executed. In addition, the mutual coupling or direct coupling or communication connection shown or discussed may be indirect coupling or communication connection through some interfaces, devices or units, or may be electrical, mechanical or other forms of connection.

[0149] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of the embodiments of the present invention, depending on actual needs.

[0150] Furthermore, the functional units in the various embodiments of this invention can be integrated into one processing unit, or each unit can exist physically separately, or two or more units can be integrated into one unit. The integrated units can be implemented in hardware or as software functional units. When using each module, user information is collected and stored only with the user's full authorization and in compliance with relevant laws and regulations, protecting the security and privacy of user data, and strictly prohibiting unauthorized access; data processing will be conducted within the scope stipulated by law and will not exceed the purpose and scope authorized by the user; at the same time, users have the rights to access, correct, delete, restrict processing, and refuse their personal data; and must strictly comply with applicable laws and regulations and conduct compliance reviews.

[0151] If the integrated unit is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or all or part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0152] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A smart monitoring method based on a large model, characterized in that, include: Receive data to be processed from smart terminals; The data to be processed includes preprocessed sensor data and structured prompts; Based on a pre-set system prompt word template, input data is generated according to pre-processed sensor data and structured prompt words; The input data is fed into a large model for parsing and reasoning to generate structured output data; Alarm information is sent to users and / or external systems based on the structured output data.

2. The intelligent monitoring method based on a large model as described in claim 1, characterized in that, The process of generating preprocessed sensor data is as follows: The application on the smart terminal receives monitoring request prompts and configuration data input by the user; the configuration data is the user's configuration of the monitoring scenario type and urgency level. The monitoring requirement prompts are processed to obtain structured prompts. The structured prompts include the monitored object, monitoring conditions, threshold parameters, duration, and urgency level of the scenario; Receive user authorization for the use of the sensor and obtain authorization information; Sensor data is collected based on a preset sampling strategy, and the sensor data is preprocessed.

3. The intelligent monitoring method based on a large model as described in claim 2, characterized in that, Sensor data is collected based on a preset sampling strategy, specifically: Based on the structured prompts and the authorization information, a sampling strategy for the corresponding sensor is determined; the sampling strategy includes sampling frequency and / or data upload strategy. Sensor data from at least one sensor is collected based on the sampling strategy described above; the sensor includes a camera, a microphone, a temperature sensor, a humidity sensor, and a gyroscope sensor.

4. The intelligent monitoring method based on a large model as described in claim 1, characterized in that, The structured output data shall include at least one of the following: indication of whether the monitoring conditions are met, event type, confidence level, and cause description.

5. The intelligent monitoring method based on a large model as described in claim 1, characterized in that, Based on the structured output data, alarm information is sent to the user and / or external systems, specifically as follows: The structured output data is parsed and judged. When the alarm triggering conditions are met, alarm information is sent to the user and / or external system through an alarm method according to the pre-configured multi-channel reminder strategy and reminder priority. The alarm method includes at least one of SMS, email, in-app push notification, smart terminal vibration, smart terminal speaker prompt sound and API interface.

6. A large-scale intelligent monitoring system, comprising an intelligent terminal and a server; characterized in that, The smart terminal is used to: collect data to be processed and send the data to be processed to the server; the data to be processed includes preprocessed sensor data and structured prompt words; The server is used for: Receive the data to be processed; Based on a pre-set system prompt word template, input data is generated according to pre-processed sensor data and structured prompt words; The input data is fed into a large model for parsing and reasoning to generate structured output data; Alarm information is sent to users and / or external systems based on the structured output data.

7. The intelligent monitoring system based on a large model as described in claim 6, characterized in that, The smart terminal includes: The user interface is used to receive monitoring request prompts and configuration data input by the user; the configuration data is the user's configuration of the monitoring scenario type and urgency level. The standardization processing unit is used to standardize the monitoring requirement prompts to obtain structured prompts; the structured prompts include the monitoring object, monitoring conditions, threshold parameters, duration, and scene urgency. The authorization unit is used to receive user authorization for the use of the sensor and obtain authorization information. The data acquisition unit is used to acquire sensor data based on a preset sampling strategy; The preprocessing unit is used to preprocess the sensor data, including compression, sampling, filtering, format conversion, and time alignment. The data transmission unit is used to send the preprocessed sensor data and structured prompts to the server.

8. The intelligent monitoring system based on a large model as described in claim 7, characterized in that, The data acquisition unit is specifically used for: Based on the structured prompts and the authorization information, a sampling strategy for the corresponding sensor is determined; the sampling strategy includes sampling frequency and / or data upload strategy. Sensor data from at least one sensor is collected based on the sampling strategy described above; the sensor includes a camera, a microphone, a temperature sensor, a humidity sensor, and a gyroscope sensor.

9. The intelligent monitoring system based on a large model as described in claim 8, characterized in that, The server includes: A data receiving unit is used to receive data to be processed from multiple smart terminals and to classify and store the data to be processed. The data processing unit is used to generate input data based on pre-processed sensor data and structured prompts, according to a pre-set system prompt template. The large model invocation unit is used to invoke the large model to parse and reason about the input data to obtain structured output data; the structured output data includes at least one of the following: indication information on whether the monitoring conditions are met, event type, confidence level, and cause explanation.

10. The intelligent monitoring system based on a large model as described in claim 9, characterized in that, The server also includes an alarm unit for: The structured output data is parsed and judged. When the alarm triggering conditions are met, alarm information is sent to the user and / or external system through an alarm method according to the pre-configured multi-channel reminder strategy and reminder priority. The alarm method includes at least one of SMS, email, in-app push notification, smart terminal vibration, smart terminal speaker prompt sound and API interface.

Images