CN117061546A - Human living space monitoring and analyzing system and method - Google Patents

Abstract

The invention provides a human living space monitoring and analyzing system and method, comprising an Internet of things sensor, image recognition equipment, voice recognition equipment and environment monitoring equipment, wherein various information in a building is monitored through the Internet of things sensor, the information obtained by the image recognition equipment, the voice recognition equipment and the Internet of things sensor is gathered in a database to realize omnibearing monitoring, the monitoring and analyzing system adopts a novel improved weighted K-means clustering algorithm, and the monitoring and analyzing system adopts various intelligent technologies including technologies of image recognition, voice recognition, internet of things and the like, so that various information in the building can be monitored and mastered in an omnibearing way, the intelligent degree, the management efficiency and the safety of the building are effectively improved, and conference content is extracted through an automatic voice recognition technology to assist management personnel to master the conference content quickly; environmental information in the building can be monitored, and safety of personnel and equipment is guaranteed.

Description

Human living space monitoring and analyzing system and method

Technical Field

The invention relates to the field of space management, in particular to a human living space monitoring and analyzing system and method.

Background

The human-occupied space building is internally provided with a large number of different devices and rooms, so that the safety can be improved by monitoring and managing the internal space of the building, and the safety of internal personnel is ensured.

In the prior art, equipment of a building in a living space is managed only through a monitoring mode, management still mainly depends on manual inspection, so that management efficiency is low, vehicles which correspondingly enter and exit can be identified and monitored, but automatic identification and monitoring are difficult to achieve for pedestrians in the past, information of different persons and the positions of the different persons cannot be timely acquired and updated, so that the equipment has larger hysteresis for position communication among the persons, the persons or the vehicles cannot be tracked under special scenes, the generated illegal behaviors are difficult to be efficiently explored, and monitoring dead angles are easy to exist.

Disclosure of Invention

Aiming at the defects existing in the prior art, the invention aims to provide a human-occupied space monitoring and analyzing system and method for solving the problems in the background art, the invention can monitor and control various information in a building in an all-round way, effectively improve the intelligent degree, management efficiency and safety of the building, effectively improve the monitoring precision and discover the illegal behavior of the building, assist management personnel to quickly master the content of meeting information, and guarantee the safety of personnel and equipment through intelligent analysis.

In order to achieve the above object, the present invention is realized by the following technical scheme: the utility model provides a human-computer space monitoring analysis system, includes thing networking sensor, image recognition equipment, speech recognition equipment and environmental monitoring equipment, monitors various information in the building through thing networking sensor, the information collection that image recognition equipment, speech recognition equipment and thing networking sensor obtained is in a database, realizes the omnidirectional monitoring, and this monitoring analysis system adopts novel modified weighting K-mean value clustering algorithm.

Further, the image recognition device uses a face recognition technology to perform identity authentication on personnel in the building, and monitors and tracks personnel, vehicles and the like.

Furthermore, the voice recognition equipment can perform all-weather voice monitoring on the internal conference and the early warning information, and semantic extraction is performed on conference contents through an automatic voice recognition technology.

Furthermore, the environment monitoring equipment can monitor the environment information in the building, including temperature, humidity, light and the like, and ensure the safety of personnel and equipment through intelligent analysis.

Furthermore, the novel improved weighted K-means clustering algorithm is optimized by using a genetic algorithm, and the specific algorithm process is as follows:

s1, initializing a population, putting k class centers into individual genes, and distributing a classified initial mark for each external data point;

s2, calculating a fitness function, adopting a weighted error square sum (weighted sum of squared errors) as the fitness function, wherein the goal is to minimize the error sum, and simultaneously considering the weight distribution of each category so as to ensure a better clustering effect;

s3, carrying out evolution on the population, generating a new generation of population through basic genetic operations such as selection, crossing, mutation and the like, and repeating the step 2 and the step 3 until a preset stopping condition is reached;

s4, selecting optimal individuals from the final generation population as a final clustering result.

A human living space monitoring and analyzing method, which uses the monitoring and analyzing system part, mainly comprises the following aspects: various devices in the building are monitored by using the Internet of things sensor, objects such as personnel and vehicles are identified by using an image identification technology, and the personnel is authenticated by using the face identification technology, so that the comprehensive monitoring is realized.

Furthermore, the voice recognition equipment is utilized to perform all-weather voice monitoring on the internal conference and early warning information, and semantic extraction is performed on conference contents through an automatic voice recognition technology to assist management staff in rapidly grasping conference information contents.

Further, environmental information in the building is monitored through environmental monitoring equipment, including temperature, humidity, light and the like, and safety of personnel and equipment is guaranteed through intelligent analysis.

Further, information obtained through the sensors of the Internet of things is gathered in a database and transmitted to the cloud server for data storage and management.

The invention has the beneficial effects that:

1. the human-occupied space monitoring and analyzing system uses various intelligent technologies, including technologies such as image recognition, voice recognition and the Internet of things, can monitor and control various information in a building in an omnibearing manner, and effectively improves the intelligent degree, management efficiency and safety of the building.

2. The human living space monitoring and analyzing system monitors various devices in a building by using the Internet of things sensor, and simultaneously identifies objects such as personnel and vehicles by using an image identification technology, and the identity of the personnel is authenticated by using a face identification technology. The multi-mode data aggregation can effectively improve the monitoring precision and discover the illegal behavior of the building; the voice recognition equipment is provided, so that all-weather voice monitoring can be carried out on the internal conference and early warning information, semantic extraction is carried out on conference contents through an automatic voice recognition technology, and management staff is assisted to quickly master the conference information contents; environmental information in the building can be monitored, including temperature, humidity, light and the like, and safety of personnel and equipment is guaranteed through intelligent analysis.

3. The human living space monitoring and analyzing system and the method apply the multi-mode technology to the field of building monitoring, so that the monitoring precision is improved, and the management efficiency and the safety are also improved; the improved weighted genetic K-means clustering algorithm provides a comprehensive, efficient, stable and plastic data classification solution, is suitable for a large number of actual data analysis and clustering scenes, and has superior performance and practicability which are incomparable with the current common clustering algorithm.

Drawings

FIG. 1 is a diagram of a human living space monitoring and analyzing system according to the present invention;

fig. 2 is a schematic diagram of a method for monitoring and analyzing a living space according to the present invention.

Detailed Description

The invention is further described in connection with the following detailed description, in order to make the technical means, the creation characteristics, the achievement of the purpose and the effect of the invention easy to understand.

Referring to fig. 1 to 2, the present invention provides a technical solution: the system comprises an Internet of things sensor, image recognition equipment, voice recognition equipment and environment monitoring equipment, wherein various information in a building is monitored through the Internet of things sensor, and the method comprises the following specific implementation steps:

step 1:

1.1 in order to implement the multi-modal intelligent building monitoring system, this embodiment requires the selection of a series of suitable devices and careful deployment and installation within the building. Such devices include, but are not limited to: the intelligent camera comprises an Internet of things sensor, an intelligent camera, indoor positioning equipment, voice recognition equipment, other environment sensors and the like.

1.2 every room within a building should be equipped with an internet of things sensor. These sensors can monitor indoor temperature, humidity, dust, carbon dioxide, etc. data. Through the sensor of the Internet of things and the network technology, data can be transmitted to a cloud server for data storage and management. At the same time, the availability and integrity of the monitored data needs to be guaranteed.

1.3 in the building, installation intelligent camera can bring more comprehensive personnel monitoring ability. By using high-definition image and face recognition technology, accurate multi-mode data acquisition and processing can be realized, and recognition and early warning of events such as personnel access, intrusion and the like are supported. The monitoring requirements for different areas in the building are different, and the number and placement positions of cameras are required to be customized and deployed for the different areas.

1.4 in buildings, in different floors, the installation of indoor positioning equipment can provide real-time personnel and equipment tracking information in the building. The device can perform fine positioning and tracking using bluetooth BLE and the like. Indoor positioning data can be used for various aspects such as staff working time statistics, staff management, tracking fields and the like.

The 1.5 voice recognition equipment can be used in areas such as conference rooms and halls, can realize functions such as real-time voice recognition and semantic understanding, and can automatically convert the functions into data which can be processed, so that subsequent data analysis and processing are facilitated.

1.6 other environmental sensors, such as air quality sensors, flammable gas sensors, etc., may also be deployed within a building to monitor the environment for pollutants or hazardous materials. The environment sensors are combined with the monitoring and alarming system, so that staff can be reminded in real time, and corresponding emergency measures are adopted to protect the safety and stability of the staff.

In summary, through reasonable deployment and use of the sensors and the devices, the efficiency and the practicability of the multi-mode intelligent building monitoring system can be greatly improved, and great digital improvement is brought to building management, so that more accurate and timely monitoring and management of people, things and things are better realized.

Step 2:

2.1, the information acquisition equipment such as the internet of things sensor, the intelligent camera, the indoor positioning equipment, the voice recognition equipment and the like are installed at different positions in a building. These devices will collect various monitoring information in real time, such as personnel information, environmental information, and device status.

2.2 the data acquisition equipment is used for carrying out centralized storage on the acquired data through the cloud server, so that the safety and stability of the data are ensured. The data acquisition period can be set according to actual application requirements so as to realize data acquisition of a plurality of time periods.

2.3, processing and analyzing the collected image data by adopting an image analysis and artificial intelligence algorithm to realize multiple functions of video image face recognition, identity authentication and the like, for example, different clothes can be identified in a monitoring image, and a deep convolutional neural network based on machine learning can be trained based on a large amount of labeling data so as to realize the functions of face recognition and the like of passengers. Meanwhile, the multi-target tracking technology can be utilized to realize real-time tracking of personnel and materials in a building.

2.4, collecting voices of areas such as conference rooms, halls and the like through voice recognition equipment, performing real-time voice recognition and understanding, converting voice information into processable digital data, and preparing for subsequent data analysis and modeling.

2.5, accurately tracking and positioning personnel in the building through the indoor positioning equipment, acquiring the position information and dynamic information of the personnel, and making data use for subsequent building management and security management.

2.6 the information acquired by the data acquisition equipment needs to be transmitted in real time so as to conveniently realize real-time monitoring and analysis of building states. Meanwhile, in the process of collecting data, effective control on the safety and privacy protection of data transmission is also required.

The specific implementation steps and modes of the step 2 part are as above. In summary, the data acquisition portion needs to integrate multiple data sources and process the acquired data using skills in different fields. In this step, it should be noted that the collected data should strictly control privacy and security issues, so as to ensure the security of the data.

Step 3:

and 3.1, establishing a data analysis platform for centrally storing and managing data. This data platform needs to have high availability, security and stability, and needs to be able to support a large amount of data storage and processing. At the same time, a proper data format is also needed to be selected to store the data, so that the data can be efficiently accessed and used in the data platform.

3.2 data analysis platforms require the use of advanced big data analysis techniques and artificial intelligence techniques to analyze and mine data, especially unstructured data in the data collected by the multimodal monitoring system. For example, collected text data is analyzed and processed using natural language processing techniques, pattern validation and clustering and induction analysis by modeling.

And 3.3, establishing an intelligent analysis workflow and implementing data management so as to ensure the accuracy and the well-being of data processing. For example, a data cleansing flow is established to perform data deduplication, missing, and correction of erroneous data to ensure the quality and accuracy of the data. At the same time, techniques for monitoring prediction, anomaly detection, etc. should be established, as well as handling performance problems that may occur in large data volumes.

3.4 the data analysis platform should adjust the environment in the building by establishing a prediction model and a decision model, so as to improve the utilization efficiency of resources. For example, optimizing the reliability of devices within a building by predicting the best actions that should be taken; the safety and the guarantee of the building are ensured by establishing a proper security mechanism.

3.5 the data analysis platform is controlled by security and privacy protection, and a high-level data science community is formed for the works. And establishing a special security mechanism and an auditing program to avoid the problems of data leakage and malicious input and ensure the security and the integrity of the data.

The implementation step of the step 3 is implemented by establishing a data analysis platform and using machine learning, data mining and other technologies, the data analysis platform is expected to provide high-quality data analysis service for building management and solid support for data-driven decision making. Meanwhile, the data analysis platform can also be used as an innovation and development platform for building management, and can be used for carrying out deep transformation and intelligent upgrading on building management.

Step 4:

4.1 building artificial intelligence algorithms and machine learning models through which data mining and analysis is performed to facilitate data driven decisions. The algorithm can be used for realizing the automatic management, preventive maintenance and other works of the building. For example, building equipment is fault detected using machine learning algorithms, enabling maintenance personnel to more accurately predict which equipment is most likely to be problematic, thereby reducing building downtime and operating costs.

And 4.2, establishing an intelligent analysis and decision-making working platform to realize intelligent management of the building. The platform needs to integrate artificial intelligence, big data and the Internet of things technology, establishes a complete digital platform and provides more practical and efficient decision support for building management.

4.3 using algorithms and models to automatically process and optimize equipment, manpower, and resources within the building. For example, management and scheduling of staff attendance scheduling may be achieved through an intelligent scheduling system and a human resource decision-making workstation. The method is characterized in that relatively strict technology is distributed according to the needs, and efficient decision is needed, so that efficient and on-site real-time personnel dispatch operation is achieved by processing data collected from various devices, and a reliable technical means is provided for stable operation and security management of a building; it is of course also possible to suggest retrofitting old equipment to reduce the capital expenditure of a building through the use of cost and benefit analyses.

4.4, building self-adaptive mechanism is built by deep learning technology, so as to realize instant monitoring, real-time processing and emergency processing of various event conditions. For example, the artificial intelligence technology can be used for monitoring parameters such as temperature, humidity, air quality and the like in a building in real time, and automatically giving an alarm when certain parameters reach a certain threat level so as to inform related personnel to take effective measures.

And 4.5, an intelligent security mechanism is established, and equipment such as the Internet of things equipment, the intelligent camera and the voice recognition equipment are used as data sources for security decision, so that real-time monitoring, processing and scheduling of building security are realized. For example, when a security system detects a potential threat, the system may automatically issue an early warning informing security personnel to handle and deal with in time.

The above is the implementation step of step 4, and by using advanced artificial intelligence and deep learning algorithms, the data driven decision will become a completely new building management mode.

Step 5:

5.1 a big data center based on cloud technology is built, all operations and data are stored in the center and are stored and shared securely. The required equipment needs to realize the instant transmission of the data to the cloud server so as to ensure the real-time performance and accuracy of the data.

5.2 constructing a data monitoring and reporting mechanism, wherein the mechanism periodically generates and reports the data monitored in real time according to actual requirements. The monitored time intervals and frequencies are determined based on the monitored metrics and business requirements while ensuring easy understanding and usability of the data output.

5.3 financial and cost benefit analysis is performed to assess the benefits of the multi-modal intelligent building monitoring system. Comparing the cost of operating the system with the benefits of use and determining the necessary adjustments to reduce costs.

And 5.4, implementing a technical risk management plan, and minimizing risks of data disclosure and privacy problems by implementing security and privacy protection measures. And measuring and calculating the influence range of the multi-mode intelligent building monitoring system in building business, and adopting authorization and access control measures to ensure the efficiency and safety of high-level and sensitive information. Meanwhile, other technical security measures, such as network security experience, forced access control, signature verification and the like, are adopted to ensure the security of the system.

5.5 build collaboration with building management teams and provide training and support. After being on line, necessary training including data operation, vulnerability diagnosis and the like is provided for business personnel and maintenance personnel so as to ensure the efficient and stable operation of the multi-mode intelligent building monitoring system.

The implementation step of the step 5 is that the multi-mode intelligent building monitoring system needs to establish a complete, efficient and stable monitoring mechanism and a data analysis platform, and update and improve according to the needs.

The embodiment adopts a novel improved weighted K-means clustering algorithm, and optimizes by utilizing a genetic algorithm, so that a better clustering result is realized. The specific algorithm process is as follows:

1. the population is initialized, k class centers are placed in the individual genes, and each external data point is assigned a classified initial marker.

2. And calculating the fitness function. The weighted error sum squared is used as a fitness function with the goal of minimizing the error sum. And meanwhile, the weight distribution of each category is considered, so that better clustering effect is ensured.

3. The population is evolved. Generating a new generation population through basic genetic operations such as selection, crossover, mutation and the like, and repeating the step 2 and the step 3 until a preset stopping condition is reached.

4. And selecting optimal individuals from the final generation population as a final clustering result.

The formula of the genetic algorithm is as follows:

-fitness function:

$F＝\frac{1}{k}\sum_{i＝1}^k\sum_{j＝1}^{n_i}w_{i,j}||x_{i,j}-c_i||^2$

where k represents the number of clusters, $n_i$ represents the number of data points in the ith cluster, $x_ { i, j } $ represents the jth data point, $c_i$ represents the cluster center of the ith cluster, $w_ { i, j } $ is a weight.

-genetic manipulation:

selection operation: suitable chromosomes are selected according to the fitness function and copied into new populations for crossover and mutation of the next generation.

Crossover operation: two individuals are randomly selected from the selected chromosomes, and their genes are crossed, one crossing point (which may be the entire gene string) is randomly selected, and the corresponding genes of the two individuals are swapped.

Mutation operation: for newly generated populations, mutation operations are performed on certain genes according to a certain probability, such as randomly changing the values of certain genes or adding or deleting certain genes.

The algorithm flow and the formula are specific improvements in the weighted genetic K-means clustering algorithm, so that the algorithm can finish the task of data clustering more stably and efficiently. It is worth noting that the randomness in the algorithm is large, so that proper technical means and optimization strategies are required to ensure that more reliable and excellent clustering results are obtained.

In this embodiment, in the weighted genetic K-means clustering algorithm, some specific designs are required for fitness functions and genetic operations:

-fitness function: in designing the fitness function, considering the influence of the distance and the weight between the data points, we can take the form of weighted error square sum, take the weighted square of the distance as the error, and add the weight coefficient to represent the importance of different samples. Therefore, the cluster to which the data point belongs can be determined more accurately, and a better clustering effect is realized.

-genetic manipulation: in genetic manipulation, selection operators need to take into account the problems of maintaining diversity of populations and avoiding premature. The crossover operator needs to ensure that the variation of the chromosome is small on the basis of variation, so that the damage to excellent gene combinations is avoided. Mutation operators control mutation probabilities to prevent populations from falling into local optima prematurely.

Through the specific design, in the clustering analysis of the data, the weighted genetic K-means clustering algorithm can more accurately perform the data clustering analysis, and a perfect clustering effect is achieved.

The invention carries out specific design on the fitness function and genetic operation in the weighted genetic K-means clustering algorithm, and can further improve the accuracy and reliability of the clustering result. The specific steps and formulas are as follows:

1. design of fitness function

In the genetic K-means clustering algorithm, the fitness function is usually calculated by adopting the error square sum. To further increase accuracy, it is necessary to design fitness functions, introduce sample weights when calculating distance, and important data points should have higher weights.

The fitness function may be designed to:

$F＝\frac{1}{\sum_{i＝1}^{k}|S_i|}\sum_{i＝1}^{k}\sum_{x\in S_{i}}w_d(x,c_i)^pw_h(x,i,p)$

where $ k represents the number of clusters, $s_i $ is the dataset of the $ i $ th cluster, $i $ is the size of the $ i $ th cluster, $c_i $ represents the center of the $ i th cluster, $w_d (x, c_i) $ is the distance between the $ x$ sample point and the center point of the $ i th cluster, $w_h (x, i, p) $ is an exponential function of $ [ x\in s_i ] $, where when $x belongs to the $i th cluster, $1, otherwise a value between $ [0,1] $ and p is an exponent.

Therefore, the fitness function takes the importance of distance and the effect of sample weight into consideration, and more accurately reflects the classification effect.

2. Design of genetic manipulation

To ensure the effectiveness and high efficiency of genetic algorithms, optimization is required on three basic operations of crossover, mutation and selection.

-a selection operation: selecting $N $individuals with the greatest fitness from the population as parents by adopting a tournament selection mode so as to ensure the combined action of diversity and fitness in the population.

-a crossover operation: and performing single-point crossover operation, namely randomly selecting two individuals, randomly selecting a position in chromosomes of the two individuals, dividing the positions of the two individuals by the position for crossover operation, and exchanging gene fragments of the two individuals.

-mutation operation: the perturbation type mutation is adopted, namely, gaussian perturbation is introduced during mutation, and the amplitude of the perturbation is controlled, so that the chromosome structure is ensured not to be completely changed on the basis of mutation.

Through the specific design, the genetic K-means clustering algorithm can analyze and process data more accurately, and more excellent clustering results are obtained.

Aiming at a weighted K-means clustering algorithm, the invention provides an improved fitness function design method so as to adapt to practical problems and improve clustering effect. The method comprises the following specific steps:

1. determining sample weights (or weight intervals): in practical problems, different data points may have different importance, representativeness, and classification difficulty, so different weight values (or weight intervals) may be determined for different data points. For example, for abnormal data points in the financial data analysis, higher weight values may be assigned to emphasize their important impact on the clustering results.

2. Calculating the distance according to the sample weight: in calculating the distance between the sample point and the cluster center, the differences of each attribute may be weighted summed taking into account the sample weights. Specifically, for the distance $d_ { ij } $ between the $i$ th sample and the $j$ th cluster center, one can calculate as follows:

$d_{ij}＝\sqrt{\sum_{m＝1}^{n}(w_{im}d_{ijm})^2}$

where $w_ { im } $ is the weight of the $ i $ sample on the $ m $ th attribute, $d_ { ijm } $ is the difference between the $ i $ sample and the $ j $ th cluster center on the $ m $ th attribute, and $ n is the number of attributes.

3. Calculating a fitness function according to the sum of squares of errors: the weighted sum of squares of errors may be used as a fitness function. Specifically, for a cluster analysis of a set of data points $c= \ { c_ {1}, c_ {2},..:

$F(C)＝\frac{\sum_{i＝1}^{k}\omega_{i}

E(C_{i})}{\sum_{i＝1}^{k}\omega_{i}|C_{i}|}$

where $ (C_ { i }) $ is the sum of the squares of the errors of the $ i $ th cluster, $\omega_ { i $ is the weight of the $ i $ th cluster, and $C_ { i } | is the number of data points in the $ i $ th cluster. The equation can ensure the coordination between the sample weight and the error square sum, thereby better reflecting the quality of the clustering result.

Through the steps, the improved fitness function can better adapt to practical problems, and more accurate clustering results are realized by considering factors such as sample weight, distance, error and the like.

In this embodiment, a specific design is performed for genetic operations in the weighted genetic K-means clustering algorithm, so as to further improve the clustering effect of the algorithm. Specifically, the invention adopts the following strategies and formulas to carry out genetic operation design:

1. selection operation

In order to ensure diversity of population and avoid the limitation of algorithm to local optimal solution, the invention adopts tournament selection operator to select operation. Specifically, the tournament selection operator has the following operational steps:

-randomly selecting $n $individuals from a population.

And selecting the individual with the highest fitness function value from the selected individuals as a parent to perform crossover and mutation operations.

The larger the random number of individuals, $n$, the easier the population will remain to be diversified, but at the same time the convergence speed of the algorithm may be correspondingly reduced.

2. Crossover operation

The interleaving operation typically employs a single point interleaving operator, which randomly selects two individuals for interleaving. The choice of crossing points is also generally random, in order to avoid the occurrence of repeated crossing points. The crossover operation can be expressed by the following formula:

-randomly selecting two genotypes.

-randomly selecting one intersection for the crossover operation.

After crossing, two new offspring are generated and added to the population.

Since the superior chromosome structure may be destroyed during crossover, crossover mutation strategies may be introduced during crossover to avoid this.

3. Mutation operation

In order to maintain the diversity of the population, the invention adopts a Gaussian mutation operator to carry out mutation operation. The Gaussian mutation operator randomly perturbs the original gene value to achieve the purpose of destroying the chromosome. The gaussian mutation operator can be expressed as:

adding a Gaussian perturbation ($\Delta_g$) to each gene of each individual.

The randomly selected heuristic factor $\Delta_g$ conforms to a Gaussian distribution.

The probability of a mutation occurring can be expressed in $p_m$.

In the mutation strategy, $p_m$ represents the probability of chromosomal mutation. The value of $ p_m $ is typically small to ensure population convergence and diversity.

In summary, the embodiment adopts a specific genetic operation design, so that the weighted genetic K-means clustering algorithm is more excellent and stable in clustering effect. In the practical application process, proper parameter adjustment is needed according to specific problems, so that the algorithm can better process data and obtain more excellent clustering results.

While the fundamental and principal features of the invention and advantages of the invention have been shown and described, it will be apparent to those skilled in the art that the invention is not limited to the details of the foregoing exemplary embodiments, but may be embodied in other specific forms without departing from the spirit or essential characteristics thereof. The present embodiments are, therefore, to be considered in all respects as illustrative and not restrictive, the scope of the invention being indicated by the appended claims rather than by the foregoing description, and all changes which come within the meaning and range of equivalency of the claims are therefore intended to be embraced therein. Any reference sign in a claim should not be construed as limiting the claim concerned.

Furthermore, it should be understood that although the present disclosure describes embodiments, not every embodiment is provided with a separate embodiment, and that this description is provided for clarity only, and that the disclosure is not limited to the embodiments described in detail below, and that the embodiments described in the examples may be combined as appropriate to form other embodiments that will be apparent to those skilled in the art.

Claims (9)

1. The utility model provides a human living space monitoring analysis system which characterized in that: the system comprises an Internet of things sensor, image recognition equipment, voice recognition equipment and environment monitoring equipment, wherein various information in a building is monitored through the Internet of things sensor, the information obtained by the image recognition equipment, the voice recognition equipment and the Internet of things sensor is integrated in a database, so that omnibearing monitoring is realized, and the monitoring analysis system adopts a novel improved weighted K-means clustering algorithm.

2. A human living space monitoring and analyzing system according to claim 1, wherein: the image recognition device uses a face recognition technology to carry out identity authentication on personnel in a building, and monitors and tracks personnel, vehicles and the like.

3. A human living space monitoring and analyzing system according to claim 2, wherein: the voice recognition equipment can perform all-weather voice monitoring on the internal conference and early warning information, and semantic extraction is performed on conference contents through an automatic voice recognition technology.

4. A human-occupied space monitoring and analyzing system according to claims 2 and 3, wherein: the environment monitoring equipment can monitor environment information in the building, including temperature, humidity, light and the like, and ensure the safety of personnel and equipment through intelligent analysis.

5. A human living space monitoring and analyzing system according to claim 1, wherein: the novel improved weighted K-means clustering algorithm is optimized by utilizing a genetic algorithm, and the specific algorithm process is as follows:

s1, initializing a population, putting k class centers into individual genes, and distributing a classified initial mark for each external data point;

s2, calculating a fitness function, adopting a weighted error square sum (weighted sum of squared errors) as the fitness function, wherein the goal is to minimize the error sum, and simultaneously considering the weight distribution of each category so as to ensure a better clustering effect;

s3, carrying out evolution on the population, generating a new generation of population through basic genetic operations such as selection, crossing, mutation and the like, and repeating the step 2 and the step 3 until a preset stopping condition is reached;

s4, selecting optimal individuals from the final generation population as a final clustering result.

6. A human living space monitoring and analyzing method, characterized in that the monitoring and analyzing method uses the monitoring and analyzing system as claimed in claim 1, and mainly comprises the following aspects: various devices in the building are monitored by using the Internet of things sensor, objects such as personnel and vehicles are identified by using an image identification technology, and the personnel is authenticated by using the face identification technology, so that the comprehensive monitoring is realized.

7. The human living space monitoring and analyzing method according to claim 6, wherein: the voice recognition equipment is utilized to perform all-weather voice monitoring on the internal conference and early warning information, and the automatic voice recognition technology is utilized to perform semantic extraction on conference contents so as to assist management personnel to quickly grasp the conference information contents.

8. The human living space monitoring and analyzing method according to claim 7, wherein: environmental information in the building is monitored through environmental monitoring equipment, including temperature, humidity, light and the like, and safety of personnel and equipment is guaranteed through intelligent analysis.

9. The human living space monitoring and analyzing method according to claim 8, wherein: information obtained through the sensors of the Internet of things is gathered in a database and transmitted to a cloud server for data storage and management.