Search

CN-122024405-A - Indoor behavior analysis and safety monitoring method and system based on multi-source perception

CN122024405ACN 122024405 ACN122024405 ACN 122024405ACN-122024405-A

Abstract

The invention belongs to the technical field of intelligent Internet of things, and particularly discloses an indoor behavior analysis and safety monitoring method and system based on multi-source perception, which can realize accurate identification and monitoring of specific daily behaviors of a user through multi-source data fusion and custom portrait matching, effectively distinguish normal behaviors from abnormal conditions and greatly reduce monitoring false alarm rate; the method combines the item early warning rules and the weighted statistical rules, can enable the monitoring early warning to have clear basis, improves the credibility and the treatment efficiency of early warning, only analyzes insensitive water, electricity, gas, heat data, does not relate to privacy information such as faces, videos, biological characteristics and the like, combines hierarchical authority management, protects the privacy of users to the greatest extent while guaranteeing the safety, does not need to wear any equipment by the users, does not need to change the existing indoor layout, can realize all-weather noninductive safety monitoring of the users, and can effectively reduce the monitoring pressure of a monitoring party while reducing the direct and indirect monitoring cost.

Inventors

  • LI GUOLIANG

Assignees

  • 深圳市慧创未来科技有限公司

Dates

Publication Date
20260512
Application Date
20260410

Claims (10)

  1. 1. The indoor behavior analysis and safety monitoring method based on multi-source perception is characterized by comprising the following steps of: Acquiring an indoor water, electricity and gas heat use data set of a user, which is transmitted by an edge gateway in real time, wherein the indoor water, electricity and gas heat use data set of the user comprises indoor water use data, electricity use data, gas use data and heat use data of the user; Determining the behavior characteristics corresponding to each daily behavior of the user according to the indoor water, electricity, gas and heat use data set of the user; constructing a daily behavior portrait of the user based on the behavior characteristics corresponding to each daily behavior of the user; The custom behavior portraits of the users are called, the daily behavior portraits of the users are matched with the custom behavior portraits, and the behavior statistical similarity is determined; when the behavior statistical similarity does not meet the set similarity threshold condition, judging that the daily behavior of the user is comprehensively abnormal, and generating corresponding first abnormal early warning information; And sending the first abnormal early warning information to a monitoring management end of the user for behavior statistics early warning.
  2. 2. The method for analyzing and monitoring indoor behaviors based on multi-source perception according to claim 1, wherein the behavior characteristics comprise water, electricity, gas, heat consumption, behavior times, behavior duration and behavior occurrence time, and the custom behavior representation comprises custom behaviors of the user and behavior characteristics corresponding to the custom behaviors.
  3. 3. The multisource awareness based indoor behavior analysis and safety monitoring method according to claim 2, wherein the method further comprises: comparing the behavior characteristics of each daily behavior of the user with a preset behavior characteristic abnormality judgment rule to judge whether the water, electricity, gas and heat consumption, the behavior times, the behavior duration and/or the behavior occurrence time of each daily behavior of the user are abnormal; confirming that the characteristics of the daily behaviors of the user are abnormal when judging that the water, electricity and gas heat consumption, the behavior times, the behavior duration and/or the behavior occurrence time of the corresponding daily behaviors of the user are abnormal, and generating corresponding second abnormal early warning information; And sending the second abnormal early warning information to a monitoring management end of the user for behavior characteristic early warning.
  4. 4. The multi-source perception based indoor behavior analysis and safety monitoring method according to claim 3, further comprising: When judging that the user has no corresponding daily behaviors beyond the continuous set time, confirming that the user has no behavior abnormality, and generating corresponding third abnormality early warning information; and sending the third abnormal early warning information to a monitoring management end of the user for non-behavior early warning.
  5. 5. The multisource awareness based indoor behavior analysis and safety monitoring method according to claim 4, further comprising: After the first abnormal early warning information is sent to the monitoring management end of the user, when the fact that the daily behavior comprehensive abnormality of the user does not exist is judged in the follow-up set number of monitoring periods, behavior statistics early warning relieving information is generated and sent to the monitoring management end of the user; after the second abnormal early warning information is sent to the monitoring management end of the user, when no daily behavior characteristic abnormality of the user is judged in a follow-up set number of monitoring periods, behavior characteristic early warning canceling information is generated and sent to the monitoring management end of the user; after the third abnormal early warning information is sent to the monitoring management end of the user, generating the non-behavior early warning release information and sending the non-behavior early warning release information to the monitoring management end of the user when the fact that the non-behavior abnormality of the user exists is judged to exist in the follow-up set number of monitoring periods.
  6. 6. The multisource awareness based indoor behavior analysis and safety monitoring method of claim 5, further comprising: importing the daily behavior portraits of the users and preset personal health behavior management rules of the users into a health knowledge base for health analysis to obtain daily behavior health analysis data of the users; Generating a monitoring period report of the user based on the daily behavior image of the user in the current monitoring period, and the first abnormal early warning information, the second abnormal early warning information and/or the third abnormal early warning information; and carrying out privacy authority grading and grading preservation on the indoor water, electricity, gas and heat use data set of the user, daily behavior images of the user, daily behavior health analysis data of the user and the monitoring period report of the user in the current monitoring period.
  7. 7. The method for analyzing and monitoring indoor behaviors based on multi-source perception according to claim 2, wherein the matching the daily behavior portraits and the habit behavior portraits of the user to determine the statistical similarity of behaviors comprises: And determining the weight corresponding to each behavior feature dimension, matching the daily behavior portrait with the habit behavior portrait of the user, and calculating the behavior statistical similarity of the daily behavior and the habit behavior at the corresponding moment by adopting a weighted summation mode according to the weight corresponding to each behavior feature dimension and the behavior feature of the daily behavior and the behavior feature of the habit behavior at the corresponding moment.
  8. 8. The method for analyzing and monitoring indoor behaviors based on multi-source sensing according to claim 1, wherein the determining each daily behavior of the user according to the indoor water, electricity, gas and heat usage data set of the user comprises: The method comprises the steps that indoor water consumption data, electricity consumption data, gas consumption data and heat consumption data of a user are matched with set daily behavior judging rules, each daily behavior of the user is determined, the daily behavior judging rules comprise a plurality of daily behaviors and water consumption data, electricity consumption data, gas consumption data and/or heat consumption data association relations corresponding to each daily behavior, the water consumption data comprise water consumption time and water consumption, the electricity consumption data comprise electricity consumption time, electricity consumption type and electricity consumption, the gas consumption data comprise gas consumption time and gas consumption, and the heat consumption data comprise heat consumption time and heat consumption.
  9. 9. The indoor behavior analysis and safety monitoring system based on multi-source perception is characterized by comprising an intelligent device layer, an edge gateway and a cloud server, wherein the intelligent device layer is used for acquiring indoor water consumption data, electricity consumption data, gas consumption data and heat consumption data of a user, the intelligent device layer comprises an intelligent circuit breaker, an intelligent water meter, an intelligent gas meter, an intelligent electric meter, an intelligent heat meter and an intelligent control valve, the edge gateway is used for acquiring the intelligent device layer in a timing or frequency conversion mode to acquire the water consumption data, the electricity consumption data, the gas consumption data and the heat consumption data of the user at corresponding indoor moments, carrying out edge synchronization processing, utilizing the synchronized water consumption data, the synchronized electricity consumption data, the synchronized gas consumption data and the synchronized heat consumption data to form an indoor water power and gas heat consumption data set of the user, and transmitting the indoor water power and the indoor gas heat consumption data set of the user to the cloud server, and the cloud server is used for executing the indoor behavior analysis and safety monitoring method based on multi-source perception.
  10. 10. Indoor behavior analysis and safety monitoring system based on multisource perception, which is characterized by comprising: a memory for storing instructions; the processor is used for reading the instructions stored in the memory and executing the indoor behavior analysis and safety monitoring method based on multi-source perception according to the instructions.

Description

Indoor behavior analysis and safety monitoring method and system based on multi-source perception Technical Field The invention belongs to the technical field of intelligent Internet of things, and particularly relates to an indoor behavior analysis and safety monitoring method and system based on multi-source perception. Background As society is accelerated to age, the number of solitary and empty-nest old people is continuously increased, and daily safety monitoring of the solitary and empty-nest old people is a focus of social attention. Meanwhile, with the development of the internet of things technology, intelligent devices such as intelligent water meters, intelligent electric meters, intelligent gas meters, intelligent circuit breakers and intelligent heat meters are gradually popularized, and a foundation is provided for acquiring water, electricity, gas and heat usage data of residents in real time. At present, monitoring for solitary and empty-nest old people mainly depends on a camera monitoring mode, a wearable device monitoring mode, a living water, electricity and gas single threshold value warning system and manual regular inspection modes. These approaches suffer from the following disadvantages: 1. The camera monitoring has the defects of high installation and maintenance cost, limited indoor coverage range, easiness in causing privacy problems and the like. 2. The wearing equipment is monitored, the equipment purchasing and maintaining cost is high, the old is depended on active wearing and charging, the actual compliance is poor, and the long-term effective use is difficult. 3. The domestic water, electricity and gas single threshold alarm system only carries out overrun alarm by setting a fixed threshold of single data (such as water consumption and electricity power), can not distinguish normal behavior from real abnormality, has extremely high false alarm rate, can not trace specific behavior, and has poor monitoring effect. 4. Manual regular inspection is limited by human resources, especially the vast land and people in rural areas, and comprehensive and timely coverage is difficult to realize. Therefore, there is a need for a home security monitoring scheme that is unobserved, does not violate privacy, is low in cost, and can effectively identify the daily activities of the monitored person and perform reliable anomaly early warning. ‌ ‌ A Disclosure of Invention The invention aims to provide an indoor behavior analysis and safety monitoring method and system based on multi-source perception, which are used for solving the problems in the prior art. In order to achieve the above purpose, the present invention adopts the following technical scheme: In a first aspect, there is provided a method for indoor behavior analysis and safety monitoring based on multi-source perception, comprising: Acquiring an indoor water, electricity and gas heat use data set of a user, which is transmitted by an edge gateway in real time, wherein the indoor water, electricity and gas heat use data set of the user comprises indoor water use data, electricity use data, gas use data and heat use data of the user; Determining the behavior characteristics corresponding to each daily behavior of the user according to the indoor water, electricity, gas and heat use data set of the user; constructing a daily behavior portrait of the user based on the behavior characteristics corresponding to each daily behavior of the user; The custom behavior portraits of the users are called, the daily behavior portraits of the users are matched with the custom behavior portraits, and the behavior statistical similarity is determined; when the behavior statistical similarity does not meet the set similarity threshold condition, judging that the daily behavior of the user is comprehensively abnormal, and generating corresponding first abnormal early warning information; And sending the first abnormal early warning information to a monitoring management end of the user for behavior statistics early warning. In one possible design, the behavior features include water, electricity, gas, heat usage, number of behaviors, duration of behaviors, and time of occurrence of behaviors, and the custom behavior representation includes each custom behavior of the user and a behavior feature corresponding to each custom behavior. In one possible design, the method further comprises: comparing the behavior characteristics of each daily behavior of the user with a preset behavior characteristic abnormality judgment rule to judge whether the water, electricity, gas and heat consumption, the behavior times, the behavior duration and/or the behavior occurrence time of each daily behavior of the user are abnormal; confirming that the characteristics of the daily behaviors of the user are abnormal when judging that the water, electricity and gas heat consumption, the behavior times, the behavior duration and/or the behavior occurrence time of the corr