CN117132945A - Multi-source data fusion-based key area security method and device - Google Patents
Multi-source data fusion-based key area security method and device Download PDFInfo
- Publication number
- CN117132945A CN117132945A CN202311390684.6A CN202311390684A CN117132945A CN 117132945 A CN117132945 A CN 117132945A CN 202311390684 A CN202311390684 A CN 202311390684A CN 117132945 A CN117132945 A CN 117132945A
- Authority
- CN
- China
- Prior art keywords
- real
- feasible solution
- security
- key area
- image data
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Granted
Links
- 238000000034 method Methods 0.000 title claims abstract description 49
- 230000004927 fusion Effects 0.000 title claims abstract description 24
- 238000012544 monitoring process Methods 0.000 claims abstract description 89
- 238000012549 training Methods 0.000 claims abstract description 67
- 238000012545 processing Methods 0.000 claims abstract description 20
- 230000008569 process Effects 0.000 claims abstract description 9
- 239000000779 smoke Substances 0.000 claims description 46
- 230000005856 abnormality Effects 0.000 claims description 16
- 230000002159 abnormal effect Effects 0.000 claims description 13
- 238000013528 artificial neural network Methods 0.000 claims description 13
- 230000007246 mechanism Effects 0.000 claims description 12
- 238000007781 pre-processing Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 3
- 230000009467 reduction Effects 0.000 claims description 3
- 230000000694 effects Effects 0.000 abstract description 13
- 241000208340 Araliaceae Species 0.000 description 1
- 235000005035 Panax pseudoginseng ssp. pseudoginseng Nutrition 0.000 description 1
- 235000003140 Panax quinquefolius Nutrition 0.000 description 1
- 230000006978 adaptation Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000013527 convolutional neural network Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 235000008434 ginseng Nutrition 0.000 description 1
- 238000011478 gradient descent method Methods 0.000 description 1
- 238000012804 iterative process Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 230000002269 spontaneous effect Effects 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/50—Context or environment of the image
- G06V20/52—Surveillance or monitoring of activities, e.g. for recognising suspicious objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Information and communication technology [ICT] specially adapted for implementation of business processes of specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/20—Image preprocessing
- G06V10/36—Applying a local operator, i.e. means to operate on image points situated in the vicinity of a given point; Non-linear local filtering operations, e.g. median filtering
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
- G06V10/77—Processing image or video features in feature spaces; using data integration or data reduction, e.g. principal component analysis [PCA] or independent component analysis [ICA] or self-organising maps [SOM]; Blind source separation
- G06V10/80—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level
- G06V10/803—Fusion, i.e. combining data from various sources at the sensor level, preprocessing level, feature extraction level or classification level of input or preprocessed data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V40/00—Recognition of biometric, human-related or animal-related patterns in image or video data
- G06V40/10—Human or animal bodies, e.g. vehicle occupants or pedestrians; Body parts, e.g. hands
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B17/00—Fire alarms; Alarms responsive to explosion
- G08B17/12—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions
- G08B17/125—Actuation by presence of radiation or particles, e.g. of infrared radiation or of ions by using a video camera to detect fire or smoke
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Multimedia (AREA)
- Business, Economics & Management (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Health & Medical Sciences (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Computing Systems (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- Tourism & Hospitality (AREA)
- Software Systems (AREA)
- Development Economics (AREA)
- Human Computer Interaction (AREA)
- Computer Security & Cryptography (AREA)
- Educational Administration (AREA)
- Nonlinear Science (AREA)
- Economics (AREA)
- Human Resources & Organizations (AREA)
- Marketing (AREA)
- Primary Health Care (AREA)
- Strategic Management (AREA)
- General Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Image Analysis (AREA)
Abstract
The invention discloses a security method and a security device for key areas based on multi-source data fusion, which belong to the technical field of security technology and data processing, and acquire fire monitoring results, personnel identity legitimacy and temperature identification results respectively by acquiring multi-source data and identifying the multi-source data, and finally acquire security identification information of the key areas according to the fire monitoring results, the personnel identity legitimacy and the temperature identification results, so that comprehensive security work of the key areas is realized, a training algorithm is provided, a related identification model is trained, global optimal values can be found in the training process, and the final training effect is ensured, thereby ensuring the identification effect.
Description
Technical Field
The invention belongs to the technical field of security and data processing, and particularly relates to a key area security method and device based on multi-source data fusion.
Background
In an actual working or living scene, there is often a need for security in a "key area" where people are concerned, to protect privacy, property and technical secrets from leakage, so as to meet working or living demands. In order to carry out security patrol on the key areas, a face recognition method is generally adopted to monitor the key areas, the problem that monitoring is not in place exists, meanwhile, a neural network is generally adopted to carry out recognition on the face recognition, if the neural network is not trained in place, the problem of false alarm can occur frequently, and therefore security significance is lost. In the prior art, a gradient descent method or a back propagation algorithm is often adopted to train the neural network, and although the training effect can be realized, the neural network is often easy to fall into a local optimal value, so that the problem of poor training effect is caused.
Disclosure of Invention
The invention provides a key area security method and device based on multi-source data fusion, which are used for solving the problems of insufficient monitoring and poor recognition effect in the prior art.
In one aspect, the invention provides a key region security method based on multi-source data fusion, which comprises the following steps:
acquiring multi-source data corresponding to various sensors in a key area, wherein the multi-source data comprises real-time image data, temperature sensing time sequence data and real-time smoke sensing data;
according to the real-time smoke sensing data, performing fire monitoring on the important areas to obtain fire monitoring results corresponding to the important areas; the fire monitoring result comprises the presence or absence of a fire;
according to the real-time image data, a pre-trained intelligent recognition model is adopted for personnel monitoring, and personnel identity legitimacy in a key area is determined, wherein the personnel identity legitimacy comprises identity legitimacy or identity non-legitimacy;
according to the temperature sensing time series data, performing temperature identification by adopting a pre-trained temperature identification model, and determining a temperature identification result corresponding to a key area, wherein the temperature identification result comprises temperature abnormality or temperature non-abnormality;
and acquiring security identification information of the key area according to the fire monitoring result, the personnel identity validity and the temperature identification result, sharing the security identification information to the cloud, and finishing security of the key area.
Further, according to the real-time smoke sensing data, performing fire monitoring on the important area to obtain a fire monitoring result corresponding to the important area, including:
judging whether the real-time smoke sensing data is empty or not, if yes, determining that a fire monitoring result corresponding to a key area is fire, otherwise, performing secondary judgment on the real-time smoke sensing data;
judging whether the real-time smoke sensing data is larger than a set smoke sensing threshold value, if so, determining that a fire disaster exists in a fire disaster monitoring result corresponding to a key area, otherwise, determining that the fire disaster does not exist in the fire disaster monitoring result corresponding to the key area, and generating smoke abnormal information;
and the smoke abnormality information is shared to terminal equipment of a manager in real time.
Further, according to the real-time image data, a pre-trained intelligent recognition model is adopted to monitor personnel, and the identity legitimacy of the personnel in the key area is determined, including:
preprocessing the real-time image data to obtain preprocessed real-time image data;
adopting a pre-trained intelligent recognition model to recognize the preprocessed real-time image data to obtain a portrait recognition result in the real-time image data;
and determining the validity of the personnel identity in the key area according to the personnel image recognition result in the real-time image data.
Further, preprocessing the real-time image data to obtain preprocessed real-time image data, including:
filtering the real-time image data to obtain real-time image data after primary processing;
graying is carried out on the real-time image data after primary processing to obtain real-time image data after secondary processing;
performing PCA dimension reduction on the real-time image data after the secondary processing to obtain real-time image data after the tertiary processing;
and converting the real-time image data after three times of processing into a vector form to obtain the real-time image data after preprocessing.
Further, the method for acquiring the pre-trained intelligent recognition model comprises the following steps:
constructing an intelligent recognition model by adopting a classified neural network, and initializing super parameters of the intelligent recognition model;
generating a plurality of feasible solutions in a solution space based on the super-parameters of the intelligent recognition model, and setting individual training times of each feasible solution and total training times of the whole training process;
judging whether the current total training times is smaller than a first set threshold value, if so, performing global traversal search on each feasible solution according to a greedy updating mechanism to obtain a feasible solution after one update, otherwise, performing local traversal search on each feasible solution according to the greedy updating mechanism to obtain a feasible solution after one update;
determining the selected probability of each feasible solution, and determining the feasible solution to be updated secondarily by adopting a roulette manner according to the selected probability of the feasible solution;
searching the feasible solution to be updated secondarily in a field searching mode to obtain a feasible solution after updating secondarily;
for the feasible solution after primary updating and the feasible solution after secondary updating, determining whether the adaptability of the feasible solution exceeds the individual training times without increasing, if so, randomly initializing the feasible solution without increasing, otherwise, judging the iteration ending condition;
judging whether the current training times is greater than the total training times, if so, outputting a feasible solution with the maximum current fitness, taking the outputted feasible solution as a final super parameter of the intelligent recognition model, and otherwise, returning to the steps of local traversal search and global traversal search.
Further, generating a plurality of feasible solutions in a solution space based on the hyper-parameters of the smart recognition model, including:
wherein,represent the firstiFirst of the feasible solutionsjThe dimensional parameters of the object are defined by the dimensions,i=1,2,…,I，Irepresenting the total number of all possible solutions,j=1,2,…,J，Jrepresenting the total number of dimensions of the feasible solution; when i=1, _a->Indicating the first of the hyper-parameters for the initialization of the smart recognition modeljDimension parameter->Representing the circumference ratio>Represent the firsti+1 th possible solutionjDimensional parameters, arcsin, represent an arcsine function.
Further, performing global traversal search on each feasible solution according to a greedy update mechanism to obtain a feasible solution after one update, including:
the global traversal search specifically includes:
wherein,represent the firsttIn the training processiFeasible solution, ->Representing updated feasible solutions +.>，Represent the firsttUpdate amount during secondary training +.>Represent the firstt-update amount during 1 training, < >>Representing inertial weights，/>Represent the firsttGlobally optimal solution in secondary training process->Represent the firstiUpdate parameters corresponding to the feasible solutions, +.>Representing the upper limit of the update parameter->Representing the update parameter lower limit->Representing the random number between (0, 1),Trepresenting a preset maximum update number;
judging whether the adaptability of the feasible solution after the global traversal search is increased, if so, accepting the global traversal search to obtain a feasible solution after one update, otherwise, rejecting the global traversal search to obtain a feasible solution after one update;
performing local traversal search on each feasible solution according to a greedy update mechanism to obtain a feasible solution after one update, wherein the method comprises the following steps:
the local traversal search is:
wherein,represent the firsttIn the training processiFeasible solution, ->Representing updated feasible solutions +.>，Representing the update coefficient between (0, 1, ">Represent the firsttIn the course of secondary training and->Different possible solutions;
judging whether the adaptability of the feasible solution after the local search is increased, if so, accepting the local search to obtain the feasible solution after one update, otherwise, rejecting the local search to obtain the feasible solution after one update.
Further, determining a selected probability of each feasible solution, and determining a secondary updated feasible solution by adopting a roulette manner according to the selected probability of the feasible solution, wherein the method comprises the following steps:
the probability of being selected for each feasible solution is determined as:
wherein,represent the firstiProbability of being selected for the individual feasible solutions, +.>Represent the firstiThe intermediate parameters corresponding to the respective feasible solutions,represent the firstiAdaptation of the individual feasible solutions, +.>Representing the function of the absolute value of the equation,Brepresenting a threshold value;
determining a feasible solution to be updated secondarily by adopting a roulette manner according to the selected probability of the feasible solution;
searching the feasible solution to be updated secondarily by adopting a field searching mode to obtain the feasible solution after being updated secondarily, wherein the method comprises the following steps:
wherein,historical optimal value representing a feasible solution to be updated twice, < >>Represent the firsttFeasible solutions to be updated secondarily determined in the course of secondary training, < >>Representing updated->，/>Represent the firsttGlobally optimal solution in secondary training process->Representing the random number in-1 or 1.
Further, according to the fire monitoring result, the personnel identity validity and the temperature identification result, security identification information of the key area is obtained, and the security identification information is shared in the cloud, comprising:
judging whether the fire monitoring result is fire, if yes, generating unsafe mark information, otherwise, judging the validity of personnel identity;
judging whether the identity legitimacy of the personnel in the key area is illegal, if so, generating unsafe mark information, otherwise, judging a temperature identification result;
judging whether the temperature identification result in the key area is abnormal, if so, generating unsafe mark information, otherwise, generating safe mark information;
transmitting the mark information to a preset appointed terminal device, and forming associated data by the mark information, time-to-time current time information, a fire monitoring result, personnel identity legitimacy and a temperature identification result to obtain security identification information of a key area;
and sharing the security identification information in the cloud by adopting an asymmetric encryption mode.
On the other hand, the invention provides a key area security device based on multi-source data fusion, which comprises a multi-source data acquisition module, a fire monitoring module, a personnel monitoring module, a temperature monitoring module and a security identification module;
the multi-source data acquisition module is used for acquiring multi-source data corresponding to various sensors in a key area, wherein the multi-source data comprises real-time image data, temperature sensing time sequence data and real-time smoke sensing data;
the fire monitoring module is used for monitoring fire in the important areas according to the real-time smoke sensing data and obtaining fire monitoring results corresponding to the important areas; the fire monitoring result comprises the presence or absence of a fire;
the personnel monitoring module is used for monitoring personnel by adopting a pre-trained intelligent recognition model according to the real-time image data, and determining personnel identity legitimacy in a key area, wherein the personnel identity legitimacy comprises identity legitimacy or identity non-legitimacy;
the temperature monitoring module is used for carrying out temperature identification by adopting a pre-trained temperature identification model according to the temperature sensing time sequence data, and determining a temperature identification result corresponding to the key area, wherein the temperature identification result comprises temperature abnormality or temperature non-abnormality;
the security identification module is used for acquiring security identification information of the key area according to the fire monitoring result, the personnel identity validity and the temperature identification result, sharing the security identification information to the cloud, and finishing security of the key area.
According to the key area security method and device based on multi-source data fusion, the multi-source data are collected and identified, the fire monitoring result, the personnel identity legitimacy and the temperature identification result are respectively obtained, and finally the security identification information of the key area is obtained according to the fire monitoring result, the personnel identity legitimacy and the temperature identification result, so that comprehensive security work of the key area is realized, a training algorithm is provided, the related identification model is trained, the global optimal value can be found in the training process, the final training effect is ensured, and the identification effect is ensured.
Drawings
The accompanying drawings, which are incorporated in and constitute a part of this specification, illustrate embodiments consistent with the invention and together with the description, serve to explain the principles of the invention.
Fig. 1 is a flowchart of a key area security method based on multi-source data fusion provided by an embodiment of the present invention.
Fig. 2 is a schematic structural diagram of a key area security device based on multi-source data fusion according to an embodiment of the present invention.
The system comprises a 201-multisource data acquisition module, a 202-fire monitoring module, a 203-personnel monitoring module, a 204-temperature monitoring module and a 205-security identification module.
Specific embodiments of the present invention have been shown by way of the above drawings and will be described in more detail below. The drawings and the written description are not intended to limit the scope of the inventive concepts in any way, but rather to illustrate the inventive concepts to those skilled in the art by reference to the specific embodiments.
Detailed Description
Reference will now be made in detail to exemplary embodiments, examples of which are illustrated in the accompanying drawings. When the following description refers to the accompanying drawings, the same numbers in different drawings refer to the same or similar elements, unless otherwise indicated. The implementations described in the following exemplary examples do not represent all implementations consistent with the invention. Rather, they are merely examples of apparatus and methods consistent with aspects of the invention as detailed in the accompanying claims.
Embodiments of the present invention are described in detail below with reference to the accompanying drawings.
Example 1
As shown in fig. 1, this embodiment provides a key area security method based on multi-source data fusion, including:
s101, multi-source data corresponding to various sensors in a key area are acquired, wherein the multi-source data comprise real-time image data, temperature sensing time sequence data and real-time smoke sensing data.
The key areas need to monitor the access of personnel in real time, so that real-time image data need to be acquired for identification, and a plurality of cameras can be arranged to monitor different directions, so that the omnibearing monitoring is realized. Electrical devices may exist in the key area, so that temperature sensing time series data needs to be collected to monitor whether the devices are operating normally, and meanwhile, the temperature sensing time series data can also be used as a fire judgment basis. In order to more accurately monitor fires in a heavy spot area, it is therefore desirable to sense real-time smoke sensing data, which, once exceeding an alarm threshold, may be considered to have occurred.
It should be noted that, in addition to the real-time image data, the temperature sensing time series data, and the real-time smoke sensing data, the multi-source data may also include other data, and the temperature sensing time series data may also transpose the infrared image to identify the temperature in the key region. For example: humidity, gas concentration and the like can be detected to realize more comprehensive monitoring, and the data of the humidity, the gas concentration and the like can be judged by using a threshold value, namely, abnormal judgment can be judged when the data exceeds the threshold value, so that the embodiment is not repeated.
And S102, carrying out fire monitoring on the important areas according to the real-time smoke sensing data, and obtaining fire monitoring results corresponding to the important areas. The fire monitoring result includes the presence or absence of a fire.
Optionally, whether a flame exists can be identified through the assistance of the real-time image data, and the fire monitoring can be realized more accurately by matching with the identification result of the real-time smoke sensing data.
And S103, monitoring personnel by adopting a pre-trained intelligent recognition model according to the real-time image data, and determining personnel identity legitimacy in the key area, wherein the personnel identity legitimacy comprises identity legitimacy or identity illegality.
By combining the description, the flame recognition model can be trained in advance, and the real-time image data can be recognized through the flame recognition model, so that the recognition result of the real-time smoke sensing data can be assisted to be recognized more accurately.
The output data of the pre-trained intelligent recognition model can be the identity unlawful category and the category of the entered personnel, so that the identity legitimacy of the personnel in the key area can be determined. In order to accurately perform face recognition, the face can be cut before recognition, so that the recognition accuracy can be improved, and the data dimension can be reduced.
S104, carrying out temperature identification by adopting a pre-trained temperature identification model according to the temperature sensing time sequence data, and determining a temperature identification result corresponding to the key area, wherein the temperature identification result comprises temperature abnormality or temperature non-abnormality.
Temperature sensing time series data corresponding to a plurality of temperature sensors can be obtained, the temperature sensors can be arranged on the electrical equipment, and when the frequent temperature of the electrical equipment is abnormal, the electrical equipment can be considered to be possibly damaged or spontaneous combustion, and the fire risk exists. When the temperature sensor is not arranged on the electrical equipment, the occurrence of temperature change indicates that high-temperature or low-temperature objects appear around, fire disaster or other abnormal phenomena possibly exist, and timely alarm is needed.
The training process of the pre-trained temperature recognition model can be the same as that of the intelligent recognition model, but the training data of the pre-trained temperature recognition model are historical temperature sensing time series data, the label data corresponding to the training data are temperature recognition results corresponding to the historical temperature sensing time series data, and the training data can be temperature recognition results recorded manually or marked manually.
S105, acquiring security identification information of the key areas according to the fire monitoring result, the personnel identity validity and the temperature identification result, sharing the security identification information on the cloud, and finishing security of the key areas.
According to the fire monitoring result, the personnel identity validity and the temperature identification result, the security situation of the key area can be comprehensively judged, and whether an alarm needs to be sent or not can be determined according to the security situation.
Optionally, in order to ensure the security of the data, the security identification information may be encrypted and then stored on the cloud.
In this embodiment, according to the real-time smoke sensing data, performing fire monitoring on a heavy area to obtain a fire monitoring result corresponding to the heavy area, including:
and judging whether the real-time smoke sensing data is empty, if so, determining that the fire monitoring result corresponding to the key area is fire, otherwise, performing secondary judgment on the real-time smoke sensing data.
When smoke is not detected, the real-time smoke sensing data may be set to null or zero. When the real-time smoke sensing data is set to zero, judging whether the real-time smoke sensing data is zero or not.
Judging whether the real-time smoke sensing data is larger than a set smoke sensing threshold value, if yes, determining that a fire disaster exists in a fire disaster monitoring result corresponding to the key area, otherwise, determining that the fire disaster does not exist in the fire disaster monitoring result corresponding to the key area, and generating smoke abnormal information. And the smoke abnormality information is shared to terminal equipment of a manager in real time.
When the real-time smoke sensing data is smaller than the set smoke sensing threshold value, people can smoke or other smoke, so that smoke abnormality information is generated and workers are prompted.
In this embodiment, according to the real-time image data, a pre-trained intelligent recognition model is adopted to monitor personnel, and determining personnel identity legitimacy in a key area includes:
and preprocessing the real-time image data to obtain the preprocessed real-time image data.
And identifying the preprocessed real-time image data by adopting a pre-trained intelligent identification model to obtain a portrait identification result in the real-time image data.
Training data of the pre-trained smart recognition model may include: the face data of the person and the corresponding identity class label are recorded, and the face data of the person and the corresponding identity class label are not recorded. It is worth to say that the identity class labels corresponding to the face data of all the non-input people are the same, so that the dimension of the output data is reduced. The identity class labels corresponding to the face data of the entered person are different, so that the identity of the person entering the key area can be identified.
And determining the validity of the personnel identity in the key area according to the personnel image recognition result in the real-time image data.
The validity of the identity of a person in a key area can be determined by: when the person image recognition result in the real-time image data is the person image of the entered person, the person identity validity can be determined to be the identity validity, otherwise, the person identity validity is determined to be the identity illegality.
In this embodiment, preprocessing the real-time image data to obtain the preprocessed real-time image data includes:
and filtering the real-time image data to obtain the real-time image data after primary processing.
And graying the real-time image data after the primary processing to obtain the real-time image data after the secondary processing.
And performing PCA dimension reduction on the real-time image data after the secondary processing to obtain the real-time image data after the tertiary processing.
And converting the real-time image data after three times of processing into a vector form to obtain the real-time image data after preprocessing.
Optionally, besides the preprocessing method, the face area can be tracked and cut, so that the face can be recognized conveniently and better.
In this embodiment, the method for obtaining the pre-trained intelligent recognition model includes:
and constructing an intelligent recognition model by using the classified neural network, and initializing super parameters of the intelligent recognition model.
Alternatively, the smart identification model may be constructed using a classification neural network including an input layer, an hidden layer, and an output layer, for example, a BP (back propagation) neural network may be used to construct the smart identification model.
It should be noted that the above-mentioned classification neural network is merely an example in the present embodiment, and other classification neural networks, such as a convolutional neural network, may also be employed. When different classification neural networks are employed, the input needs to be converted into a corresponding form to ensure that the classification neural network can accurately identify the data.
Based on the super parameters of the intelligent recognition model, generating a plurality of feasible solutions in a solution space, and setting individual training times of each feasible solution and total training times of the whole training process.
When the number of exercises exceeds the total number of exercises for the whole training process, the iterative process may be ended. When the fitness of the feasible solution exceeds the number of individual training times without increasing, the feasible solution can be considered to be worse, and the feasible solution generated can be discarded or mutated so as to improve the training effect.
And judging whether the current total training times is smaller than a first set threshold value, if so, performing global traversal search on each feasible solution according to a greedy updating mechanism to obtain a feasible solution after one update, otherwise, performing local traversal search on each feasible solution according to the greedy updating mechanism to obtain a feasible solution after one update.
Through the steps, global traversal search can be performed in the early stage, and local search can be performed in the later stage, so that a better training effect is achieved.
And determining the selected probability of each feasible solution, and determining the feasible solution to be secondarily updated by adopting a roulette manner according to the selected probability of the feasible solution.
And searching the feasible solution to be updated secondarily in a field searching mode to obtain the feasible solution after updating secondarily.
When the adaptability of a feasible solution is higher, the better the position of the feasible solution is proved, the higher the probability of being selected is, and the updating speed is improved by carrying out secondary updating on the selected feasible solution.
Alternatively, the fitness may be obtained by taking the reciprocal of the error function value, and to avoid zero denominator, the error function value may be added to a very small constant (e.g., 0.00001) and then taken to obtain the fitness.
For the feasible solution after the primary updating and the feasible solution after the secondary updating, determining whether the adaptability of the solution exceeds the individual training times without increasing, if so, carrying out random initialization on the feasible solution without increasing, otherwise, carrying out judgment on iteration ending conditions.
Judging whether the current training times is greater than the total training times, if so, outputting a feasible solution with the maximum current fitness, taking the outputted feasible solution as a final super parameter of the intelligent recognition model, and otherwise, returning to the steps of local traversal search and global traversal search.
Optionally, an fitness threshold may also be set, and when the fitness of the global optimal value is greater than the fitness threshold, the training process may be ended.
In this embodiment, generating a plurality of feasible solutions in a solution space based on the hyper-parameters of the smart recognition model includes:
wherein,represent the firstiFirst of the feasible solutionsjThe dimensional parameters of the object are defined by the dimensions,i=1,2,…,I，Irepresenting the total number of all possible solutions,j=1,2,…,J，Jrepresenting the total number of dimensions of feasible solutions. When i=1, _a->Indicating the first of the hyper-parameters for the initialization of the smart recognition modeljWeishen (Ginseng radix)Count (n)/(l)>Representing the circumference ratio>Represent the firsti+1 th possible solutionjDimensional parameters, arcsin, represent an arcsine function.
By the generation method, feasible solutions can be distributed more uniformly in the solution space during initial generation, and the global optimal value can be found, so that the training effect is improved, and the final face recognition effect is ensured.
In this embodiment, performing global traversal search on each feasible solution according to a greedy update mechanism to obtain a feasible solution after one update, including:
the global traversal search specifically includes:
wherein,represent the firsttIn the training processiFeasible solution, ->Representing updated feasible solutions +.>，Represent the firsttUpdate amount during secondary training +.>Represent the firstt-update amount during 1 training, < >>Representing inertial weights, ++>Represent the firsttGlobally optimal solution in secondary training process->Represent the firstiThe update parameters corresponding to the respective feasible solutions,representing the upper limit of the update parameter->Representing the update parameter lower limit->Representing the random number between (0, 1),Tindicating a preset maximum number of updates.
Judging whether the adaptability of the feasible solution after the global traversal search is increased, if so, accepting the global traversal search to obtain the feasible solution after one update, otherwise, rejecting the global traversal search to obtain the feasible solution after one update.
Performing local traversal search on each feasible solution according to a greedy update mechanism to obtain a feasible solution after one update, wherein the method comprises the following steps:
the local traversal search is:
wherein,represent the firsttIn the training processiFeasible solution, ->Representing updated feasible solutions +.>，/>Representing the update coefficient between (0, 1, ">Represent the firsttIn the course of secondary training and->Different possible solutions.
Judging whether the adaptability of the feasible solution after the local search is increased, if so, accepting the local search to obtain the feasible solution after one update, otherwise, rejecting the local search to obtain the feasible solution after one update.
In this embodiment, determining the probability of being selected for each feasible solution, and determining a feasible solution for a secondary update by using a roulette manner according to the probability of being selected for the feasible solution includes:
the probability of being selected for each feasible solution is determined as:
wherein,represent the firstiProbability of being selected for the individual feasible solutions, +.>Represent the firstiIntermediate parameters corresponding to the feasible solutions，Represent the firstiAdaptation of the individual feasible solutions, +.>Representing the function of the absolute value of the equation,Brepresenting a threshold.
And determining the feasible solution to be updated secondarily by adopting a roulette manner according to the selected probability of the feasible solution.
Searching the feasible solution to be updated secondarily by adopting a field searching mode to obtain the feasible solution after being updated secondarily, wherein the method comprises the following steps:
wherein,historical optimal value representing a feasible solution to be updated twice, < >>Represent the firsttFeasible solutions to be updated secondarily determined in the course of secondary training, < >>Representing updated->，/>Represent the firsttGlobally optimal solution in secondary training process->Representing the random number in-1 or 1.
In this embodiment, according to a fire monitoring result, personnel identity validity and a temperature identification result, security identification information of a key area is obtained, and the security identification information is shared in a cloud, including:
judging whether the fire monitoring result is fire, if yes, generating unsafe mark information, otherwise, judging the validity of personnel identity.
Judging whether the identity legitimacy of the personnel in the key area is illegal, if so, generating unsafe mark information, otherwise, judging the temperature identification result.
Judging whether the temperature identification result in the key area is abnormal, if so, generating unsafe mark information, otherwise, generating safe mark information.
And transmitting the mark information to a preset appointed terminal device, and forming associated data by the mark information, the time-to-time information, the fire monitoring result, the personnel identity validity and the temperature identification result to obtain security identification information of the key area.
And sharing the security identification information in the cloud by adopting an asymmetric encryption mode. For example, RSA (Rivest-Shamir-Adleman) may be used to encrypt the security identification information, and then the encrypted data is transmitted to the cloud for storage.
Optionally, alarms with different levels can be generated according to the abnormal data quantity, such as three-level alarms generated when one of personnel identity legitimacy, a temperature identification result and a fire monitoring result is abnormal; generating a secondary alarm when two of the personnel identity legitimacy, the temperature identification result and the fire monitoring result are abnormal; when three of the personnel identity legitimacy, the temperature identification result and the fire monitoring result are abnormal, a first-level alarm is generated; thereby the security situation in the key area can be further reflected.
Example 2
As shown in fig. 2, the present embodiment provides a key area security device based on multi-source data fusion, which includes a multi-source data acquisition module 201, a fire monitoring module 202, a personnel monitoring module 203, a temperature monitoring module 204 and a security identification module 205.
The multi-source data acquisition module 201 is configured to acquire multi-source data corresponding to a plurality of sensors in a key area, where the multi-source data includes real-time image data, temperature sensing time series data, and real-time smoke sensing data.
The fire monitoring module 202 is configured to monitor fire in a heavy area according to the real-time smoke sensing data, and obtain a fire monitoring result corresponding to the heavy area. The fire monitoring result includes the presence or absence of a fire.
The personnel monitoring module 203 is configured to monitor personnel according to the real-time image data by using a pre-trained intelligent recognition model, and determine personnel identity legitimacy in the key area, where the personnel identity legitimacy includes identity legitimacy or identity non-legitimacy.
The temperature monitoring module 204 is configured to perform temperature identification by using a pre-trained temperature identification model according to the temperature sensing time series data, and determine a temperature identification result corresponding to the key area, where the temperature identification result includes temperature abnormality or temperature non-abnormality.
The security identification module 205 is configured to obtain security identification information of a key area according to a fire monitoring result, personnel identity validity and a temperature identification result, and share the security identification information with a cloud end to complete security of the key area.
The key region security device based on multi-source data fusion provided in this embodiment can execute the method embodiment described in embodiment 1, and has similar beneficial effects and principles, and is not described herein.
According to the key area security method and device based on multi-source data fusion, the multi-source data are collected and identified, the fire monitoring result, the personnel identity legitimacy and the temperature identification result are respectively obtained, and finally the security identification information of the key area is obtained according to the fire monitoring result, the personnel identity legitimacy and the temperature identification result, so that comprehensive security work of the key area is realized, a training algorithm is provided, the related identification model is trained, the global optimal value can be found in the training process, the final training effect is ensured, and the identification effect is ensured.
Other embodiments of the invention will be apparent to those skilled in the art from consideration of the specification and practice of the invention disclosed herein. This invention is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the present disclosure as come within known or customary practice within the art to which the invention pertains. It is intended that the specification and examples be considered as exemplary only, with a true scope and spirit of the invention being indicated by the following claims.
It is to be understood that the invention is not limited to the precise arrangements and instrumentalities shown in the drawings, which have been described above, and that various modifications and changes may be effected without departing from the scope thereof. The scope of the invention is limited only by the appended claims.
Claims (10)
1. The key area security method based on multi-source data fusion is characterized by comprising the following steps of:
acquiring multi-source data corresponding to various sensors in a key area, wherein the multi-source data comprises real-time image data, temperature sensing time sequence data and real-time smoke sensing data;
according to the real-time smoke sensing data, performing fire monitoring on the important areas to obtain fire monitoring results corresponding to the important areas; the fire monitoring result comprises the presence or absence of a fire;
according to the real-time image data, a pre-trained intelligent recognition model is adopted for personnel monitoring, and personnel identity legitimacy in a key area is determined, wherein the personnel identity legitimacy comprises identity legitimacy or identity non-legitimacy;
according to the temperature sensing time series data, performing temperature identification by adopting a pre-trained temperature identification model, and determining a temperature identification result corresponding to a key area, wherein the temperature identification result comprises temperature abnormality or temperature non-abnormality;
and acquiring security identification information of the key area according to the fire monitoring result, the personnel identity validity and the temperature identification result, sharing the security identification information to the cloud, and finishing security of the key area.
2. The method for security of key areas based on multi-source data fusion according to claim 1, wherein the step of performing fire monitoring on the key areas according to the real-time smoke sensing data to obtain fire monitoring results corresponding to the key areas comprises the following steps:
judging whether the real-time smoke sensing data is empty or not, if yes, determining that a fire monitoring result corresponding to a key area is fire, otherwise, performing secondary judgment on the real-time smoke sensing data;
judging whether the real-time smoke sensing data is larger than a set smoke sensing threshold value, if so, determining that a fire disaster exists in a fire disaster monitoring result corresponding to a key area, otherwise, determining that the fire disaster does not exist in the fire disaster monitoring result corresponding to the key area, and generating smoke abnormal information;
and the smoke abnormality information is shared to terminal equipment of a manager in real time.
3. The method for security of key areas based on multi-source data fusion according to claim 1, wherein the step of monitoring personnel by using a pre-trained intelligent recognition model according to the real-time image data to determine the validity of personnel identity in the key areas comprises the following steps:
preprocessing the real-time image data to obtain preprocessed real-time image data;
adopting a pre-trained intelligent recognition model to recognize the preprocessed real-time image data to obtain a portrait recognition result in the real-time image data;
and determining the validity of the personnel identity in the key area according to the personnel image recognition result in the real-time image data.
4. The method for security of key areas based on multi-source data fusion according to claim 3, wherein preprocessing the real-time image data to obtain the preprocessed real-time image data comprises:
filtering the real-time image data to obtain real-time image data after primary processing;
graying is carried out on the real-time image data after primary processing to obtain real-time image data after secondary processing;
performing PCA dimension reduction on the real-time image data after the secondary processing to obtain real-time image data after the tertiary processing;
and converting the real-time image data after three times of processing into a vector form to obtain the real-time image data after preprocessing.
5. The method for security of key areas based on multi-source data fusion according to claim 3, wherein the method for obtaining the pre-trained intelligent recognition model is as follows:
constructing an intelligent recognition model by adopting a classified neural network, and initializing super parameters of the intelligent recognition model;
generating a plurality of feasible solutions in a solution space based on the super-parameters of the intelligent recognition model, and setting individual training times of each feasible solution and total training times of the whole training process;
judging whether the current total training times is smaller than a first set threshold value, if so, performing global traversal search on each feasible solution according to a greedy updating mechanism to obtain a feasible solution after one update, otherwise, performing local traversal search on each feasible solution according to the greedy updating mechanism to obtain a feasible solution after one update;
determining the selected probability of each feasible solution, and determining the feasible solution to be updated secondarily by adopting a roulette manner according to the selected probability of the feasible solution;
searching the feasible solution to be updated secondarily in a field searching mode to obtain a feasible solution after updating secondarily;
for the feasible solution after primary updating and the feasible solution after secondary updating, determining whether the adaptability of the feasible solution exceeds the individual training times without increasing, if so, randomly initializing the feasible solution without increasing, otherwise, judging the iteration ending condition;
judging whether the current training times is greater than the total training times, if so, outputting a feasible solution with the maximum current fitness, taking the outputted feasible solution as a final super parameter of the intelligent recognition model, and otherwise, returning to the steps of local traversal search and global traversal search.
6. The method for security of key areas based on multi-source data fusion according to claim 5, wherein generating a plurality of feasible solutions in a solution space based on the hyper-parameters of the intelligent recognition model comprises:
wherein,represent the firstiFirst of the feasible solutionsjThe dimensional parameters of the object are defined by the dimensions,i=1,2,…,I，Irepresenting the total number of all possible solutions,j=1,2,…,J，Jrepresenting the total number of dimensions of the feasible solution; when i=1, _a->Indicating the first of the hyper-parameters for the initialization of the smart recognition modeljDimension parameter->Representing the circumference ratio>Represent the firsti+1 th possible solutionjDimensional parameters, arcsin, represent an arcsine function.
7. The method for security of key areas based on multi-source data fusion according to claim 6, wherein performing global traversal search on each feasible solution according to a greedy update mechanism to obtain a feasible solution after one update comprises:
the global traversal search specifically includes:
wherein,represent the firsttIn the training processiFeasible solution, ->Representing updated feasible solutions +.>，/>Represent the firsttUpdate amount during secondary training +.>Represent the firstt-update amount during 1 training, < >>Representing inertial weights, ++>Represent the firsttGlobally optimal solution in secondary training process->Represent the firstiUpdate parameters corresponding to the feasible solutions, +.>Representing the upper limit of the update parameter->Representing the update parameter lower limit->Representing the random number between (0, 1),Trepresenting a preset maximum update number;
judging whether the adaptability of the feasible solution after the global traversal search is increased, if so, accepting the global traversal search to obtain a feasible solution after one update, otherwise, rejecting the global traversal search to obtain a feasible solution after one update;
performing local traversal search on each feasible solution according to a greedy update mechanism to obtain a feasible solution after one update, wherein the method comprises the following steps:
the local traversal search is:
wherein,represent the firsttIn the training processiFeasible solution, ->Representing updated feasible solutions +.>，/>Representing the update coefficient between (0, 1, ">Represent the firsttIn the course of secondary training and->Different possible solutions;
judging whether the adaptability of the feasible solution after the local search is increased, if so, accepting the local search to obtain the feasible solution after one update, otherwise, rejecting the local search to obtain the feasible solution after one update.
8. The multi-source data fusion-based key area security method of claim 7, wherein determining the probability of being selected for each feasible solution, and determining a secondary updated feasible solution by way of roulette according to the probability of being selected for the feasible solution, comprises:
the probability of being selected for each feasible solution is determined as:
wherein,represent the firstiProbability of being selected for the individual feasible solutions, +.>Represent the firstiIntermediate parameters corresponding to the feasible solutions, +.>Represent the firstiAdaptation of the individual feasible solutions, +.>Representation calculationAn absolute value function of the absolute value,Brepresenting a threshold value;
determining a feasible solution to be updated secondarily by adopting a roulette manner according to the selected probability of the feasible solution;
searching the feasible solution to be updated secondarily by adopting a field searching mode to obtain the feasible solution after being updated secondarily, wherein the method comprises the following steps:
wherein,historical optimal value representing a feasible solution to be updated twice, < >>Represent the firsttFeasible solutions to be updated secondarily determined in the course of secondary training, < >>Representing updated->，/>Represent the firsttGlobally optimal solution in secondary training process->Representing the random number in-1 or 1.
9. The multi-source data fusion-based key area security method of any one of claims 1-8, wherein obtaining security identification information of a key area and sharing the security identification information to a cloud end according to a fire monitoring result, personnel identity validity and a temperature identification result comprises:
judging whether the fire monitoring result is fire, if yes, generating unsafe mark information, otherwise, judging the validity of personnel identity;
judging whether the identity legitimacy of the personnel in the key area is illegal, if so, generating unsafe mark information, otherwise, judging a temperature identification result;
judging whether the temperature identification result in the key area is abnormal, if so, generating unsafe mark information, otherwise, generating safe mark information;
transmitting the mark information to a preset appointed terminal device, and forming associated data by the mark information, time-to-time current time information, a fire monitoring result, personnel identity legitimacy and a temperature identification result to obtain security identification information of a key area;
and sharing the security identification information in the cloud by adopting an asymmetric encryption mode.
10. The key area security device based on multi-source data fusion is characterized by comprising a multi-source data acquisition module, a fire monitoring module, a personnel monitoring module, a temperature monitoring module and a security identification module;
the multi-source data acquisition module is used for acquiring multi-source data corresponding to various sensors in a key area, wherein the multi-source data comprises real-time image data, temperature sensing time sequence data and real-time smoke sensing data;
the fire monitoring module is used for monitoring fire in the important areas according to the real-time smoke sensing data and obtaining fire monitoring results corresponding to the important areas; the fire monitoring result comprises the presence or absence of a fire;
the personnel monitoring module is used for monitoring personnel by adopting a pre-trained intelligent recognition model according to the real-time image data, and determining personnel identity legitimacy in a key area, wherein the personnel identity legitimacy comprises identity legitimacy or identity non-legitimacy;
the temperature monitoring module is used for carrying out temperature identification by adopting a pre-trained temperature identification model according to the temperature sensing time sequence data, and determining a temperature identification result corresponding to the key area, wherein the temperature identification result comprises temperature abnormality or temperature non-abnormality;
the security identification module is used for acquiring security identification information of the key area according to the fire monitoring result, the personnel identity validity and the temperature identification result, sharing the security identification information to the cloud, and finishing security of the key area.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311390684.6A CN117132945B (en) | 2023-10-25 | 2023-10-25 | Multi-source data fusion-based key area security method and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202311390684.6A CN117132945B (en) | 2023-10-25 | 2023-10-25 | Multi-source data fusion-based key area security method and device |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN117132945A true CN117132945A (en) | 2023-11-28 |
| CN117132945B CN117132945B (en) | 2024-01-09 |
Family
ID=88856716
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202311390684.6A Active CN117132945B (en) | 2023-10-25 | 2023-10-25 | Multi-source data fusion-based key area security method and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN117132945B (en) |
Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20100049380A (en) * | 2008-11-03 | 2010-05-12 | 삼일씨티에스(주) | Method for management of building using robot and system thereof |
| CN102982008A (en) * | 2012-11-01 | 2013-03-20 | 山东大学 | Complicated function maximum and minimum solving method by means of parallel artificial bee colony algorithm based on computer cluster |
| CN106228184A (en) * | 2016-07-19 | 2016-12-14 | 东北大学 | A kind of based on the blast furnace fault detection system and the method that optimize extreme learning machine |
| CN106952442A (en) * | 2017-04-05 | 2017-07-14 | 合肥酷睿网络科技有限公司 | A kind of family safety protection system and its means of defence |
| CN108628209A (en) * | 2018-04-28 | 2018-10-09 | 黄河科技学院 | A kind of architectural engineering detection safety device |
| CN110493566A (en) * | 2019-08-05 | 2019-11-22 | 成都油管家科技有限公司 | One kind being based on video analysis Safety of Gas Station identifying system |
| CN111383409A (en) * | 2020-03-17 | 2020-07-07 | 南通感创电子科技有限公司 | Safety video monitoring device for intelligent building and use method thereof |
| CN112767624A (en) * | 2020-12-18 | 2021-05-07 | 山东盛帆蓝海电气有限公司 | Intelligent building safety joint defense system and method based on communication network |
| WO2022179266A1 (en) * | 2021-03-22 | 2022-09-01 | 中铁九局集团第四工程有限公司 | Deviation rectification control method and apparatus for shield tunneling attitude |
| CN115952824A (en) * | 2022-12-26 | 2023-04-11 | 西安交通大学 | Big data fault diagnosis method for aviation manufacturing equipment |
| CN116012780A (en) * | 2022-12-22 | 2023-04-25 | 广西电网有限责任公司河池供电局 | Fire disaster monitoring method and system based on image recognition |
| CN116030591A (en) * | 2022-12-30 | 2023-04-28 | 中国矿业大学 | Intelligent inspection alarm system and method for mine external fire disaster based on Internet of things |
| CN116246401A (en) * | 2022-12-13 | 2023-06-09 | 上海优客工场创业服务有限公司 | Monitoring system for community public area management |
| CN116362549A (en) * | 2023-05-22 | 2023-06-30 | 北京航天常兴科技发展股份有限公司 | Fire disaster prevention and control method based on data information mining technology |
| CN116524674A (en) * | 2023-03-23 | 2023-08-01 | 贵州电网有限责任公司 | Intelligent power distribution room fire early monitoring method and system based on multi-source data fusion |
| CN116612601A (en) * | 2023-05-15 | 2023-08-18 | 内蒙古大学 | Underground space's chamber security protection monitored control system |
-
2023
- 2023-10-25 CN CN202311390684.6A patent/CN117132945B/en active Active
Patent Citations (16)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR20100049380A (en) * | 2008-11-03 | 2010-05-12 | 삼일씨티에스(주) | Method for management of building using robot and system thereof |
| CN102982008A (en) * | 2012-11-01 | 2013-03-20 | 山东大学 | Complicated function maximum and minimum solving method by means of parallel artificial bee colony algorithm based on computer cluster |
| CN106228184A (en) * | 2016-07-19 | 2016-12-14 | 东北大学 | A kind of based on the blast furnace fault detection system and the method that optimize extreme learning machine |
| CN106952442A (en) * | 2017-04-05 | 2017-07-14 | 合肥酷睿网络科技有限公司 | A kind of family safety protection system and its means of defence |
| CN108628209A (en) * | 2018-04-28 | 2018-10-09 | 黄河科技学院 | A kind of architectural engineering detection safety device |
| CN110493566A (en) * | 2019-08-05 | 2019-11-22 | 成都油管家科技有限公司 | One kind being based on video analysis Safety of Gas Station identifying system |
| CN111383409A (en) * | 2020-03-17 | 2020-07-07 | 南通感创电子科技有限公司 | Safety video monitoring device for intelligent building and use method thereof |
| CN112767624A (en) * | 2020-12-18 | 2021-05-07 | 山东盛帆蓝海电气有限公司 | Intelligent building safety joint defense system and method based on communication network |
| WO2022179266A1 (en) * | 2021-03-22 | 2022-09-01 | 中铁九局集团第四工程有限公司 | Deviation rectification control method and apparatus for shield tunneling attitude |
| CN116246401A (en) * | 2022-12-13 | 2023-06-09 | 上海优客工场创业服务有限公司 | Monitoring system for community public area management |
| CN116012780A (en) * | 2022-12-22 | 2023-04-25 | 广西电网有限责任公司河池供电局 | Fire disaster monitoring method and system based on image recognition |
| CN115952824A (en) * | 2022-12-26 | 2023-04-11 | 西安交通大学 | Big data fault diagnosis method for aviation manufacturing equipment |
| CN116030591A (en) * | 2022-12-30 | 2023-04-28 | 中国矿业大学 | Intelligent inspection alarm system and method for mine external fire disaster based on Internet of things |
| CN116524674A (en) * | 2023-03-23 | 2023-08-01 | 贵州电网有限责任公司 | Intelligent power distribution room fire early monitoring method and system based on multi-source data fusion |
| CN116612601A (en) * | 2023-05-15 | 2023-08-18 | 内蒙古大学 | Underground space's chamber security protection monitored control system |
| CN116362549A (en) * | 2023-05-22 | 2023-06-30 | 北京航天常兴科技发展股份有限公司 | Fire disaster prevention and control method based on data information mining technology |
Non-Patent Citations (3)
| Title |
|---|
| FULI ZHONG 等: "A modified ABC algorithm based on improved-global-best-guided approach and adaptive-limit strategy for global optimization", 《APPLIED SOFT COMPUTING》, pages 469 - 486 * |
| 盛盟;: "基于物联网的家庭安防系统设计", 《信息记录材料》, vol. 21, no. 02, pages 205 - 206 * |
| 程磊生: "人工蜂群算法的改进及其在神经网络中的应用", 《中国优秀硕士学位论文全文数据库信息科技辑》, pages 140 - 36 * |
Also Published As
| Publication number | Publication date |
|---|---|
| CN117132945B (en) | 2024-01-09 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110087099A (en) | A kind of monitoring method and system for protecting privacy | |
| CN109980781A (en) | A kind of transformer substation intelligent monitoring system | |
| CN101496074A (en) | Device and method for detecting suspicious activity, program, and recording medium | |
| CN111414854A (en) | Target crowd trajectory tracking method and device | |
| CN112382407A (en) | Risk management and control method and device, electronic equipment and storage medium | |
| CN107766830B (en) | A kind of image detection alarm method and relevant device | |
| US20240095862A1 (en) | Method for determining dangerousness of person, apparatus, system and storage medium | |
| CN105844846B (en) | A kind of method for safety monitoring based on Ibeacon technologies, apparatus and system | |
| CN117671887B (en) | Intelligent security early warning management method and system based on big data | |
| CN110674761A (en) | Regional behavior early warning method and system | |
| CN117319451B (en) | Urban fire-fighting Internet of things supervision system based on multi-mode big data and method thereof | |
| CN114665608B (en) | Intelligent sensing inspection system and method for transformer substation | |
| CN115393566A (en) | Fault identification and early warning method and device for power equipment, storage medium and equipment | |
| CN111506710A (en) | Information sending method and device based on rumor prediction model and computer equipment | |
| CN115330129A (en) | Enterprise safety risk early warning analysis method and system | |
| CN117972687B (en) | Block chain-based intelligent power monitoring system | |
| CN118053261B (en) | Anti-spoofing early warning method, device, equipment and medium for smart campus | |
| CN117061211A (en) | Data processing method and system based on network security management | |
| CN108537920A (en) | Visitor's monitoring method based on recognition of face and system | |
| CN111523776A (en) | Smart city system and method based on illegal target risk assessment | |
| Li et al. | Dynamic risk assessment of emergency evacuation in large public buildings: A case study | |
| CN117132945B (en) | Multi-source data fusion-based key area security method and device | |
| CN113011893A (en) | Data processing method and device, computer equipment and storage medium | |
| CN114783097A (en) | Hospital epidemic prevention management system and method | |
| CN114446487A (en) | Body temperature detection and trajectory tracking method based on face recognition |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |