CN116863399A - Network security monitoring system and method based on artificial intelligence - Google Patents
Network security monitoring system and method based on artificial intelligence Download PDFInfo
- Publication number
- CN116863399A CN116863399A CN202310820143.6A CN202310820143A CN116863399A CN 116863399 A CN116863399 A CN 116863399A CN 202310820143 A CN202310820143 A CN 202310820143A CN 116863399 A CN116863399 A CN 116863399A
- Authority
- CN
- China
- Prior art keywords
- data
- video data
- personnel
- behavior
- person
- 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.)
- Pending
Links
- 238000000034 method Methods 0.000 title claims abstract description 59
- 238000012544 monitoring process Methods 0.000 title claims abstract description 27
- 238000013473 artificial intelligence Methods 0.000 title claims abstract description 26
- 206010000117 Abnormal behaviour Diseases 0.000 claims abstract description 50
- 230000002159 abnormal effect Effects 0.000 claims abstract description 24
- 238000012545 processing Methods 0.000 claims abstract description 24
- 238000012795 verification Methods 0.000 claims abstract description 4
- 230000006399 behavior Effects 0.000 claims description 118
- 238000007405 data analysis Methods 0.000 claims description 6
- 238000006243 chemical reaction Methods 0.000 claims description 5
- 230000008569 process Effects 0.000 claims description 5
- 238000003062 neural network model Methods 0.000 claims description 4
- 238000012549 training Methods 0.000 claims description 4
- 238000001514 detection method Methods 0.000 claims description 3
- 238000003708 edge detection Methods 0.000 claims description 3
- 230000009471 action Effects 0.000 description 4
- 238000010586 diagram Methods 0.000 description 2
- 230000003278 mimic effect Effects 0.000 description 2
- 230000004048 modification Effects 0.000 description 2
- 238000012986 modification Methods 0.000 description 2
- 230000005856 abnormality Effects 0.000 description 1
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000008859 change Effects 0.000 description 1
- 230000006872 improvement Effects 0.000 description 1
- 230000003068 static 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
- G06T—IMAGE DATA PROCESSING OR GENERATION, IN GENERAL
- G06T17/00—Three dimensional [3D] modelling, e.g. data description of 3D objects
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V10/00—Arrangements for image or video recognition or understanding
- G06V10/40—Extraction of image or video features
- G06V10/44—Local feature extraction by analysis of parts of the pattern, e.g. by detecting edges, contours, loops, corners, strokes or intersections; Connectivity analysis, e.g. of connected components
-
- 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
- G06V10/7715—Feature extraction, e.g. by transforming the feature space, e.g. multi-dimensional scaling [MDS]; Mappings, e.g. subspace methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
- G06V20/00—Scenes; Scene-specific elements
- G06V20/40—Scenes; Scene-specific elements in video content
- G06V20/46—Extracting features or characteristics from the video content, e.g. video fingerprints, representative shots or key frames
-
- G—PHYSICS
- G08—SIGNALLING
- G08B—SIGNALLING OR CALLING SYSTEMS; ORDER TELEGRAPHS; ALARM SYSTEMS
- G08B31/00—Predictive alarm systems characterised by extrapolation or other computation using updated historic data
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Multimedia (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Computing Systems (AREA)
- Software Systems (AREA)
- Artificial Intelligence (AREA)
- Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Evolutionary Computation (AREA)
- General Health & Medical Sciences (AREA)
- Medical Informatics (AREA)
- Computer Graphics (AREA)
- Geometry (AREA)
- Business, Economics & Management (AREA)
- Emergency Management (AREA)
- Alarm Systems (AREA)
Abstract
The invention discloses a network security monitoring system and method based on artificial intelligence, and belongs to the technical field of computers. The system comprises a data acquisition module, a verification module, an artificial intelligent management module, a database, a safety protection module and an alarm module; the data acquisition module is used for acquiring video data in the regional network and transmitting the acquired video data to the data processing module; the data processing module is used for converting the acquired video data into a 3D model; the artificial intelligent management module is used for predicting the behavior characteristics of personnel according to the video data processing information and the stored data information; the database is used for storing data information; the mobile phone terminal is used for a manager to check the collected video data; the alarm module is used for alarming abnormal conditions; in the invention, the influence of surrounding adjacent objects on the behavior characteristics of personnel is eliminated, so that the abnormal behavior prediction of the system is more accurate.
Description
Technical Field
The invention relates to the technical field of computers, in particular to a network security monitoring system and method based on artificial intelligence.
Background
Artificial intelligence is intended to provide computers with the ability to simulate and mimic human intelligence, often using a variety of techniques and methods, to enable computers to understand, learn, infer and perform tasks to mimic human intelligence. Currently, artificial intelligence techniques are applied to a variety of fields, including intelligent security.
The intelligent security is mainly realized by utilizing an artificial intelligent system; some security systems predict abnormal behaviors by analyzing behavior feature data of people, but when the number of people is too large, the distances among people can influence the behavior features of the people, so that the predicted abnormal behaviors are inaccurate.
Disclosure of Invention
The invention aims to provide a network security monitoring system and method based on artificial intelligence, which are used for solving the problems in the background technology.
In order to solve the technical problems, the invention provides the following technical scheme: the network security monitoring system based on artificial intelligence comprises a data acquisition module, a data processing module, an artificial intelligence management module, a database, a mobile phone end and an alarm module;
the data acquisition module is used for acquiring video data in the regional network and transmitting the acquired video data to the data processing module; the data processing module is used for converting the acquired video data into a 3D model; the artificial intelligent management module is used for predicting the behavior characteristics of personnel according to the video data processing information and the stored data information; the database is used for storing data information; the mobile phone terminal is used for a manager to check the collected video data; the alarm module is used for alarming abnormal conditions;
the stored data comprises historical behavior data and historical video acquisition data.
According to the technical scheme, the data acquisition module acquires video data through the panoramic camera;
through installing panoramic camera, make the video data of gathering more complete, be convenient for the unusual action of analyst.
According to the technical scheme, the data processing module comprises a data receiving unit and a video converting unit;
the data receiving unit is used for sending the collected video data;
the video conversion unit is used for converting video data into a 3D model and recording the position information of each pixel in the model.
According to the technical scheme, the artificial intelligent management module comprises a data analysis unit, a safety monitoring unit and an intelligent control unit;
the data analysis unit is used for comparing the collected video data with historical behavior data stored in the database and identifying the behavior category of personnel;
the safety detection unit is used for analyzing abnormal behavior data of personnel;
the intelligent control unit is used for sending the abnormal behavior video of the personnel to the mobile phone end and controlling the alarm module to alarm.
According to the technical scheme, the mobile phone terminal comprises a video viewing unit and an alarm management unit;
the video viewing unit is used for viewing the collected abnormal video data and the historical video collection data in the database by a manager;
the alarm management unit is used for a manager to remotely control the alarm to send out or close an alarm.
An artificial intelligence based network security monitoring method, the method comprising the following steps:
s10, acquiring video data in a regional network, processing the acquired video data, and converting the video data into a 3D model;
s20, analyzing a 3D model corresponding to each frame of video data to obtain the shortest distance L between the person and surrounding adjacent objects min (t)；
S30, analyzing the video data, and predicting whether the behaviors of the personnel in the video data are reasonably avoided according to the historical behavior data in the database and the shortest distance between the personnel and surrounding adjacent objects in the video data; if the collision is reasonable, repeating the step S20; if the collision is unreasonable, executing a step S40;
s40, judging whether the behaviors of the personnel in the video data are abnormal according to the behaviors of the personnel in the video data and the abnormal behavior data in the database; if the behavior is normal, repeatedly executing the step S20; if the behavior is abnormal, executing the step S50;
s50, analyzing abnormal behaviors of the personnel in the video data, and predicting the behavior category of the personnel;
s60, according to the predicted personnel behavior category; when the predicted person behavior type is dangerous behavior, video data are sent to a manager to control an alarm to alarm; when the predicted person behavior category is other behaviors, repeatedly executing the step S20;
the database is used for storing historical behavior data, historical video acquisition data and abnormal behavior data; t represents a time stamp of each frame of video; the abnormal behavior data comprise dangerous behavior data and other behavior data recorded in a database; the dangerous behavior data includes various dangerous behaviors, and the dangerous behaviors can be different according to different fields.
Converting video data into a 3D model and obtaining the shortest distance L between personnel and surrounding adjacent objects min The method comprises the following steps:
s101, processing the acquired video data to obtain 3D image data of the acquired video;
s102, dispersing the 3D image into a series of volume pixels, wherein each volume pixel is a point in the 3D model;
s103, detecting edge information by using a Canny edge detection method, tracking position information of continuous edge points by adopting a tracking algorithm, and establishing a 3D model of each frame of video data by the position information of the edge points;
s104, distinguishing personnel in the acquired video from surrounding adjacent objects by using a human body detector;
s105, analyzing position information of personnel and surrounding adjacent objects in the video data in the horizontal direction, and selecting a position information set A (t) of a 3D model edge point of the personnel and a position information set B (t) of a 3D model edge point of the surrounding adjacent object on one side of the personnel, which is closer to the surrounding adjacent object, according to the relative positions of the personnel and the surrounding adjacent object;
wherein a (t) = { a 1 (t)、a 2 (t)、...、a m (t)}；a 1 (t)＝(x 1 ,y 1 ,z 1 )、a 2 (t)＝x 2 ,y 2 ,z 2 )、...、a m (t)＝(x m ,y m ,z m (；x i ，y i and zi Representing three-dimensional space coordinates of each point of a component person in the 3D model; m represents the number of coordinates of the positional information of the bounding box on the side closer to the adjacent object by the composing person; b (t) = { B 1 (t)、b 2 (t)、...、b s (t)}；b 1 (t)＝(X 1 ,Y 1 ,Z 1 (、b 2 (t)＝X 2 ,Y 2 ,z 2 (、...、b s (t)＝(X n ,Y n ,Z n )；X i ，Y i and Zi Representing three-dimensional space coordinates of each point constituting surrounding adjacent objects in the 3D model; n represents the number of coordinates constituting the positional information of the bounding box on the side of the surrounding adjacent object closer to the person;
because the video images are influenced by the video resolution, the higher the video resolution is changed due to the fact that the number of m and n is different from the acquired video resolution, the larger the number of m and n is, and the more accurate the position information of the edge points of the 3D model of the person and the surrounding adjacent objects is established; by analyzing the relative position information of the personnel and surrounding adjacent objects in the horizontal direction, the closer side between the personnel and the surrounding adjacent objects is selected as the position information of the edge points, so that the operation amount of the system is reduced;
s106, according to the set A (t) and the set B (t), respectively calculating the distance L between each spatial coordinate point of the set A (t) and each spatial coordinate point of the set B (t) in each frame of video data to obtain the shortest distance L between the person and the adjacent objects around min The method comprises the steps of carrying out a first treatment on the surface of the According to the formula:
judging whether the behavior of the person is influenced by the surrounding adjacent objects or not by calculating the shortest distance between the person and the surrounding adjacent objects; the shortest distance is calculated through the established 3D model and the tracing mode, so that the behavior prediction of the personnel is more accurate.
The method for judging whether the behaviors of the personnel in the video data are abnormal comprises the following steps:
s201, analyzing historical behavior data of personnel avoidance in a database, and corresponding to L min When (t), establishing a motion track function f (t) of the person and a motion track function g (t) of the surrounding adjacent objects in the process of avoiding the person in the same time according to the space coordinates of the person and the surrounding adjacent objects;
s202, obtaining the longest distance k between a person and surrounding adjacent objects in the same time in the historical behavior data according to functions f (t) and g (t) max And shortest distance k min ；
S203, training the longest distance and shortest distance between the person and the adjacent objects around in the same time when the person dodges in each historical behavior data by using the neural network model to obtain a shortest distance threshold K when the person dodges the adjacent objects around min And a longest distance threshold K max ；
S204, L is min (t) respectively with K min and Kmax Comparing;
s205, when K max ≥L min (t)≥K min When the person behavior in the collected video data is predicted to be reasonable avoidance, the step S204 is continuously executed at the moment; when L min (t)＜K min Or K max ＜L min When (t), predicting that the human behavior in the collected video data is unreasonable avoidance, and executing step S206;
s206, comparing the personnel behavior information in the collected video data with abnormal behavior data in a database; if the same abnormal behavior exists, the behavior of the person is abnormal; if the same abnormal behavior does not exist, the personnel act normally;
through L min (t) and K min and Kmax Comparing, the abnormal behavior characteristics of personnel caused by avoiding surrounding adjacent objects are eliminated; training the objective function through the neural network model to obtain a reasonable threshold value for avoiding the personnel, and primarily judging the behavior characteristics of the personnel, so that the judgment error caused by the behavior characteristic change caused by the avoidance of the personnel to the surrounding adjacent objects can be effectively prevented; and comparing the acquired personnel behavior information in the video data with abnormal behavior data in the database to perform secondary judgment, so that the predicted personnel behavior abnormality is more accurate.
When a person passes through a certain area, the static object does not have too much influence on the behavior characteristics of the person, and some unpredictable dynamic objects greatly influence the behavior characteristics of the person, so that the influence of surrounding dynamic adjacent objects on the abnormal behavior of the person is considered, the abnormal behavior of the person is judged more accurately by the system, and the surrounding adjacent objects which are used as judgment personnel in the collected video data are dynamic;
since the movement direction of the personnel and the surrounding adjacent objects can be the same in the collected video data, L min Always at K min and Kmax The abnormal behavior can not be judged correctly; by collecting L in video data min (t) corresponding spatial coordinate information of the person and the surrounding adjacent objects, and obtaining the movement speed and the movement direction of the person and the surrounding adjacent objects by adopting an RNN algorithm and />The method for judging the abnormal behavior of the personnel when the movement direction of the personnel and the surrounding adjacent objects is the same comprises the following steps: comparison->And->Size of the material; when->When the method is used, the personnel behavior information in the video data is directly compared with the abnormal behavior data in the database; when-> When the time T for reasonably avoiding the personnel and surrounding adjacent objects in the video data is predicted 0 The method comprises the steps of carrying out a first treatment on the surface of the According to the formula:
will L min (t)＝K max When as initial time, record L min (t) at threshold value K min and Kmax At time t of (a) 0 The method comprises the steps of carrying out a first treatment on the surface of the Let T and T 0 Comparing; when T is 0 ≥t 0 When the method is used, the personnel behaviors in the collected video data are predicted to be reasonably avoided; when T is 0 ＜t 0 When the method is used, the personnel behavior in the collected video data is predicted to be unreasonably avoided, and the personnel behavior information is directly compared with the abnormal behavior data in the database;
wherein ,representing the direction and speed of movement; the movement direction is forward, backward, left and right in the horizontal direction,representing the projection of the movement velocity in the horizontal direction and in the movement direction.
When the behavior category of the personnel is predicted to be dangerous behavior, the collected video data of the personnel are sent to a mobile phone of a manager, and an alarm is controlled to give an alarm; the manager can watch the collected abnormal person video through the mobile phone and control the alarm to continuously alarm or stop alarming through the mobile phone; the processed collected video data are stored in a database, and a manager can check the collected video data through a time stamp after login verification is successful;
the system automatically identifies suspicious personnel and sends the video data of the suspicious personnel to the mobile phone of the manager, so that the safety of the management system is improved, the manager only needs to observe the sent abnormal video data through the mobile phone to judge the safety problem in the area, and the management efficiency of the manager is improved.
Compared with the prior art, the invention has the following beneficial effects: the network security monitoring system and the network security monitoring method based on artificial intelligence are provided, the shortest distance between a person and an object is calculated by establishing a 3D model and a tracing mode, and on the premise that the influence of surrounding adjacent objects on the behavior characteristics of the person is eliminated, the behavior characteristics of the person are analyzed, so that the abnormal behavior prediction of the system is more accurate; by sending the video data of the abnormal person to the manager, the office efficiency of the manager and the security capability of the system are improved.
Drawings
The accompanying drawings are included to provide a further understanding of the invention and are incorporated in and constitute a part of this specification, illustrate the invention and together with the embodiments of the invention, serve to explain the invention. In the drawings:
FIG. 1 is a schematic diagram of an artificial intelligence based network security monitoring system of the present invention;
FIG. 2 is a schematic diagram of steps of an artificial intelligence based network security monitoring method of the present invention.
Detailed Description
The following description of the embodiments of the present invention will be made clearly and completely with reference to the accompanying drawings, in which it is apparent that the embodiments described are only some embodiments of the present invention, but not all embodiments. All other embodiments, which can be made by those skilled in the art based on the embodiments of the invention without making any inventive effort, are intended to be within the scope of the invention.
Referring to fig. 1 to 2, the present invention provides the following technical solutions: the network security monitoring system based on artificial intelligence comprises a data acquisition module, a data processing module, an artificial intelligence management module, a database, a mobile phone end and an alarm module;
the data acquisition module is used for acquiring video data in the regional network and transmitting the acquired video data to the data processing module; the data processing module is used for converting the acquired video data into a 3D model; the artificial intelligent management module is used for predicting the behavior characteristics of personnel according to the video data processing information and the stored data information; the database is used for storing data information; the mobile phone terminal is used for a manager to check the collected video data; the alarm module is used for alarming abnormal conditions;
the stored data comprises historical behavior data and historical video acquisition data.
The data acquisition module acquires video data through the panoramic camera.
The data processing module comprises a data receiving unit and a video conversion unit;
the data receiving unit is used for sending the collected video data;
the video conversion unit is used for converting video data into a 3D model and recording the position information of each pixel in the model.
The artificial intelligent management module comprises a data analysis unit, a safety monitoring unit and an intelligent control unit;
the data analysis unit is used for comparing the collected video data with historical behavior data stored in the database and identifying the behavior category of personnel;
the safety detection unit is used for analyzing abnormal behavior data of personnel;
the intelligent control unit is used for sending the abnormal behavior video of the personnel to the mobile phone end and controlling the alarm module to alarm.
The mobile phone terminal comprises a video viewing unit and an alarm management unit;
the video viewing unit is used for viewing the collected abnormal video data and the historical video collection data in the database by a manager;
the alarm management unit is used for a manager to remotely control the alarm to send out or close an alarm.
An artificial intelligence based network security monitoring method, the method comprising the following steps:
s10, acquiring video data in a regional network, processing the acquired video data, and converting the video data into a 3D model;
s20, analyzing a 3D model corresponding to each frame of video data to obtain the shortest distance L between the person and surrounding adjacent objects min (t)；
S30, analyzing the video data, and predicting whether the behaviors of the personnel in the video data are reasonably avoided according to the historical behavior data in the database and the shortest distance between the personnel and surrounding adjacent objects in the video data; if the collision is reasonable, repeating the step S20; if the collision is unreasonable, executing a step S40;
s40, judging whether the behaviors of the personnel in the video data are abnormal according to the behaviors of the personnel in the video data and the abnormal behavior data in the database; if the behavior is normal, repeatedly executing the step S20; if the behavior is abnormal, executing the step S50;
s50, analyzing abnormal behaviors of the personnel in the video data, and predicting the behavior category of the personnel;
s60, according to the predicted personnel behavior category; when the predicted person behavior type is dangerous behavior, video data are sent to a manager to control an alarm to alarm; when the predicted person behavior category is other behaviors, repeatedly executing the step S20;
the database is used for storing historical behavior data, historical video acquisition data and abnormal behavior data; t represents a time stamp of each frame of video; the abnormal behavior data includes dangerous behavior data and other behavior data recorded in a database.
Converting video data into a 3D model and obtaining the shortest distance L between personnel and surrounding adjacent objects min The method comprises the following steps:
s101, processing the acquired video data to obtain 3D image data of the acquired video;
s102, dispersing the 3D image into a series of volume pixels, wherein each volume pixel is a point in the 3D model;
s103, detecting edge information by using a Canny edge detection method, tracking position information of continuous edge points by adopting a tracking algorithm, and establishing a 3D model of each frame of video data by the position information of the edge points;
s104, distinguishing personnel in the acquired video from surrounding adjacent objects by using a human body detector;
s105, analyzing position information of personnel and surrounding adjacent objects in the video data in the horizontal direction, and selecting a position information set A (t) of a 3D model edge point of the personnel and a position information set B (t) of a 3D model edge point of the surrounding adjacent object on one side of the personnel, which is closer to the surrounding adjacent object, according to the relative positions of the personnel and the surrounding adjacent object;
wherein a (t) = { a 1 (t)、a 2 (t)、...、a m (t)}；a 1 (t)＝(x 1 ,y 1 ,z 1 )、a 2 (t)＝x 2 ,y 2 ,z 2 (、...、a m (t)＝(x m ,y m ,z m )；x i ，y i and zi Representing three-dimensional space coordinates of each point of a component person in the 3D model; m represents the number of coordinates of the positional information of the bounding box on the side closer to the adjacent object by the composing person; b (t) = { B 1 (t)、b 2 (t)、...、b s (t)}；b 1 (t)＝(X 1 ,Y 1 ,Z 1 )、b 2 (t)＝X 2 ,Y 2 ,z 2 )、...、b s (t)＝(X n ,Y n ,Z n )；X i ，Y i and Zi Representing three-dimensional space coordinates of each point constituting surrounding adjacent objects in the 3D model; n represents the number of coordinates constituting the positional information of the bounding box on the side of the surrounding adjacent object closer to the person;
s106, respectively calculating each spatial coordinate point of the set A (t) and each space of the set B (t) in each frame of video data according to the set A (t) and the set B (t)The distance L of the coordinate point is used for obtaining the shortest distance L between the person and the adjacent objects around min The method comprises the steps of carrying out a first treatment on the surface of the According to the formula:
the method for judging whether the behaviors of the personnel in the video data are abnormal comprises the following steps:
s201, analyzing historical behavior data of personnel avoidance in a database, and corresponding to L min When (t), establishing a motion track function f (t) of the person and a motion track function g (t) of the surrounding adjacent objects in the process of avoiding the person in the same time according to the space coordinates of the person and the surrounding adjacent objects;
s202, obtaining the longest distance k between a person and surrounding adjacent objects in the same time in the historical behavior data according to functions f (t) and g (t) max And shortest distance k min ；
S203, training the longest distance and shortest distance between the person and the adjacent objects around in the same time when the person dodges in each historical behavior data by using the neural network model to obtain a shortest distance threshold K when the person dodges the adjacent objects around min And a longest distance threshold K max ；
S204, L is min (t) respectively with K min and Kmax Comparing;
s205, when K max ≥L min (t)≥K min When the person behavior in the collected video data is predicted to be reasonable avoidance, the step S204 is continuously executed at the moment; when L min (t)＜K min Or K max ＜L min When (t), predicting that the human behavior in the collected video data is unreasonable avoidance, and executing step S206;
s206, comparing the personnel behavior information in the collected video data with abnormal behavior data in a database; if the same abnormal behavior exists, the behavior of the person is abnormal; if the same abnormal behavior does not exist, the personnel act normally.
Number of videos acquiredThe surrounding adjacent objects which are used as judgment staff are dynamic; by collecting L in video data min (t) corresponding spatial coordinate information of the person and the surrounding adjacent objects, and obtaining the movement speed and the movement direction of the person and the surrounding adjacent objects by adopting an RNN algorithm and />The method for judging the abnormal behavior of the personnel when the movement direction of the personnel and the surrounding adjacent objects is the same comprises the following steps: comparison->And->Size of the material; when->When the method is used, the personnel behavior information in the video data is directly compared with the abnormal behavior data in the database; when-> When the time T for reasonably avoiding the personnel and surrounding adjacent objects in the video data is predicted 0 The method comprises the steps of carrying out a first treatment on the surface of the According to the formula:
will L min (t)＝K max When as initial time, record L min (t) at threshold value K min and Kmax At time t of (a) 0 The method comprises the steps of carrying out a first treatment on the surface of the Let T and T 0 Comparing; when T is 0 ≥t 0 When the method is used, the personnel behaviors in the collected video data are predicted to be reasonably avoided; when T is 0 ＜t 0 When the method is used, the personnel behavior in the collected video data is predicted to be unreasonably avoided, and the personnel behavior information is directly compared with the abnormal behavior data in the database;
wherein ,representing the direction and speed of movement; the movement direction is forward, backward, left and right in the horizontal direction,representing the projection of the movement velocity in the horizontal direction and in the movement direction.
When the behavior category of the personnel is predicted to be dangerous behavior, the collected video data of the personnel are sent to a mobile phone of a manager, and an alarm is controlled to give an alarm; the manager can watch the collected abnormal person video through the mobile phone and control the alarm to continuously alarm or stop alarming through the mobile phone; the processed collected video data are stored in a database, and a manager can check the collected video data through a time stamp after login verification is successful.
In this embodiment:
the system is an area security video monitoring system of a stock exchange; an alarm is arranged around the area, and security personnel can control an alarm switch through a mobile phone; abnormal behaviors of the personnel in the database comprise transacting business 1, transacting business 2 and dangerous behaviors; wherein the dangerous behaviour may be theft;
acquiring video data in an area through a camera, converting the video data into a 3D model, and analyzing position information of a spatial coordinate point on one side, which is close to each other, of a person A and a surrounding person B in the video data of a set A (t) and a set B (t) in the 3D model; obtaining the shortest distance L between the person and the adjacent objects around min The method comprises the steps of carrying out a first treatment on the surface of the Analyzing historical behavior data of personnel avoidance to obtain a shortest distance threshold K when the personnel avoid surrounding adjacent objects min =0.1m, longest distance threshold K max ＝1.5m；
At a certain time t 1 Person a is behind, person B is in front,at this time, the shortest distance L between person A and person B min (t 1 )＝1.5m；K max ≥L min (t 1 )≥K min Predicting the abnormal behavior of the personnel A to reasonably avoid; the moving speed of the person A in the horizontal direction is 1.5m/s, the moving speed of the surrounding person B in the horizontal direction is 1m/s, and the moving direction is forward; at t 1 Recording time as an initial time; predicting time for reasonable avoidance of personnel and surrounding adjacent objects in video data
At t 2 At the moment, person A is in front, person B is in back, at which point L min (t 2 )＝1.5m；K max ≥L min (t 1 )≥K min At this time L min (t) at L min (t) at threshold value K min and Kmax At time t of (a) 0 ＝t 2 -t 1 ＝5s；t 0 ＜T 0 Therefore, the personnel behaviors in the collected video data are predicted to be reasonably avoided;
at another time t as personnel A and B move 3 At this time L min (t 2 )＝0.05m；L min (t 3 )＜K min The method comprises the steps of carrying out a first treatment on the surface of the The specific abnormal behavior of the person A is possibly stolen by comparing the abnormal behavior with the abnormal behavior of the person in the database, and at the moment, the artificial intelligent management module controls the alarm to alarm and sends the video data to security personnel.
It is noted that relational terms such as first and second, and the like are used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Moreover, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus.
Finally, it should be noted that: the foregoing description is only a preferred embodiment of the present invention, and the present invention is not limited thereto, but it is to be understood that modifications and equivalents of some of the technical features described in the foregoing embodiments may be made by those skilled in the art, although the present invention has been described in detail with reference to the foregoing embodiments. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention should be included in the protection scope of the present invention.
Claims (10)
1. An artificial intelligence based network security monitoring system, which is characterized in that: the system comprises a data acquisition module, a data processing module, an artificial intelligent management module, a database, a mobile phone terminal and an alarm module;
the data acquisition module is used for acquiring video data in the regional network and transmitting the acquired video data to the data processing module; the data processing module is used for converting the acquired video data into a 3D model; the artificial intelligent management module is used for predicting the behavior characteristics of personnel according to the video data processing information and the stored data information; the database is used for storing data information; the mobile phone terminal is used for a manager to check the collected video data; the alarm module is used for alarming abnormal conditions;
the stored data comprises historical behavior data and historical video acquisition data.
2. An artificial intelligence based network security monitoring system according to claim 1 wherein: the data acquisition module acquires video data through the panoramic camera.
3. An artificial intelligence based network security monitoring system according to claim 1 wherein: the data processing module comprises a data receiving unit and a video conversion unit;
the data receiving unit is used for sending the collected video data;
the video conversion unit is used for converting video data into a 3D model and recording the position information of each pixel in the model.
4. An artificial intelligence based network security monitoring system according to claim 1 wherein: the artificial intelligent management module comprises a data analysis unit, a safety monitoring unit and an intelligent control unit;
the data analysis unit is used for comparing the collected video data with historical behavior data stored in the database and identifying the behavior category of personnel;
the safety detection unit is used for analyzing abnormal behavior data of personnel;
the intelligent control unit is used for sending the abnormal behavior video of the personnel to the mobile phone end and controlling the alarm module to alarm.
5. An artificial intelligence based network security monitoring system according to claim 1 wherein: the mobile phone terminal comprises a video viewing unit and an alarm management unit;
the video viewing unit is used for viewing the collected abnormal video data and the historical video collection data in the database by a manager;
the alarm management unit is used for a manager to remotely control the alarm to send out or close an alarm.
6. A network security monitoring method based on artificial intelligence is characterized in that: the method comprises the following steps:
s10, acquiring video data in a regional network, processing the acquired video data, and converting the video data into a 3D model;
s20, analyzing a 3D model corresponding to each frame of video data to obtain the shortest distance L between the person and surrounding adjacent objects min (t)；
S30, analyzing the video data, and predicting whether the behaviors of the personnel in the video data are reasonably avoided according to the historical behavior data in the database and the shortest distance between the personnel and surrounding adjacent objects in the video data; if the collision is reasonable, repeating the step S20; if the collision is unreasonable, executing a step S40;
s40, judging whether the behaviors of the personnel in the video data are abnormal according to the behaviors of the personnel in the video data and the abnormal behavior data in the database; if the behavior is normal, repeatedly executing the step S20; if the behavior is abnormal, executing the step S50;
s50, analyzing abnormal behaviors of the personnel in the video data, and predicting the behavior category of the personnel;
s60, according to the predicted personnel behavior category; when the predicted person behavior type is dangerous behavior, video data are sent to a manager to control an alarm to alarm; when the predicted person behavior category is other behaviors, repeatedly executing the step S20;
the database is used for storing historical behavior data, historical video acquisition data and abnormal behavior data; t represents a time stamp of each frame of video; the abnormal behavior data includes dangerous behavior data and other behavior data recorded in a database.
7. The network security monitoring method based on artificial intelligence according to claim 6, wherein: converting video data into a 3D model and obtaining the shortest distance L between personnel and surrounding adjacent objects min The method comprises the following steps:
s101, processing the acquired video data to obtain 3D image data of the acquired video;
s102, dispersing the 3D image into a series of volume pixels, wherein each volume pixel is a point in the 3D model;
s103, detecting edge information by using a Canny edge detection method, tracking position information of continuous edge points by adopting a tracking algorithm, and establishing a 3D model of each frame of video data by the position information of the edge points;
s104, distinguishing personnel in the acquired video from surrounding adjacent objects by using a human body detector;
s105, analyzing position information of personnel and surrounding adjacent objects in the video data in the horizontal direction, and selecting a position information set A (t) of a 3D model edge point of the personnel and a position information set B (t) of a 3D model edge point of the surrounding adjacent object on one side of the personnel, which is closer to the surrounding adjacent object, according to the relative positions of the personnel and the surrounding adjacent object;
wherein a (t) = { a 1 (t)、a 2 (t)、...、a m (t)}；a 1 (t)＝x 1 ,y 1 ,z 1 、a 2 (t)＝x 2 ,y 2 ,z 2 、...、a m (t)＝x m ,y m ,z m ；x i ，y i and zi Representing three-dimensional space coordinates of each point of a component person in the 3D model; m represents the number of coordinates of the positional information of the bounding box on the side closer to the adjacent object by the composing person; b (t) = { B 1 (t)、b 2 (t)、...、b s (t)}；b 1 (t)＝X 1 ,Y 1 ,Z 1 、b 2 (t)＝X 2 ,Y 2 ,z 2 、...、b s (t)＝X n ,Y n ,Z n ；X i ，Y i and Zi Representing three-dimensional space coordinates of each point constituting surrounding adjacent objects in the 3D model; n represents the number of coordinates constituting the positional information of the bounding box on the side of the surrounding adjacent object closer to the person;
s106, according to the set A (t) and the set B (t), respectively calculating the distance L between each spatial coordinate point of the set A (t) and each spatial coordinate point of the set B (t) in each frame of video data to obtain the shortest distance L between the person and the adjacent objects around min The method comprises the steps of carrying out a first treatment on the surface of the According to the formula:
8. the network security monitoring method based on artificial intelligence according to claim 7, wherein: the method for judging whether the behaviors of the personnel in the video data are abnormal comprises the following steps:
s201, analyzing historical behavior data of personnel avoidance in a database, and corresponding to L min When (t), establishing a motion track function f (t) of the person and a motion track function g (t) of the surrounding adjacent objects in the process of avoiding the person in the same time according to the space coordinates of the person and the surrounding adjacent objects;
s202, obtaining the longest distance k between a person and surrounding adjacent objects in the same time in the historical behavior data according to functions f (t) and g (t) max And shortest distance k min ；
S203, training the longest distance and shortest distance between the person and the adjacent objects around in the same time when the person dodges in each historical behavior data by using the neural network model to obtain a shortest distance threshold K when the person dodges the adjacent objects around min And a longest distance threshold K max ；
S204, L is min (t) respectively with K min and Kmax Comparing;
s205, when K max ≥L min (t)≥K min When the person behavior in the collected video data is predicted to be reasonable avoidance, the step S204 is continuously executed at the moment; when L min (t)＜K min Or K max ＜L min When (t), predicting that the human behavior in the collected video data is unreasonable avoidance, and executing step S206;
s206, comparing the personnel behavior information in the collected video data with abnormal behavior data in a database; if the same abnormal behavior exists, the behavior of the person is abnormal; if the same abnormal behavior does not exist, the personnel act normally.
9. The network security monitoring method based on artificial intelligence according to claim 8, wherein: surrounding adjacent objects which are used as judgment personnel in the collected video data are dynamic; by collecting L in video data min (t) corresponding spatial coordinate information of the person and surrounding adjacent objects, and obtaining the person by adopting an RNN algorithmSpeed and direction of movement of the person and surrounding adjacent objects and />The method for judging the abnormal behavior of the personnel when the movement direction of the personnel and the surrounding adjacent objects is the same comprises the following steps: comparison->And->Size of the material; when->When the method is used, the personnel behavior information in the video data is directly compared with the abnormal behavior data in the database; when->When the time T for reasonably avoiding the personnel and surrounding adjacent objects in the video data is predicted 0 The method comprises the steps of carrying out a first treatment on the surface of the According to the formula:
will L min (t)＝K max When as initial time, record L min (t) at threshold value K min and Kmax At time t of (a) 0 The method comprises the steps of carrying out a first treatment on the surface of the Let T and T 0 Comparing; when T is 0 ≥t 0 When the method is used, the personnel behaviors in the collected video data are predicted to be reasonably avoided; when T is 0 ＜t 0 When the method is used, the personnel behavior in the collected video data is predicted to be unreasonably avoided, and the personnel behavior information is directly compared with the abnormal behavior data in the database;
wherein ,representing the direction and speed of movement; the movement direction is forward, backward, left and right in the horizontal direction, < >>Representing the projection of the movement velocity in the horizontal direction and in the movement direction.
10. The network security monitoring method based on artificial intelligence according to claim 8, wherein: when the behavior category of the personnel is predicted to be dangerous behavior, the collected video data of the personnel are sent to a mobile phone of a manager, and an alarm is controlled to give an alarm; the manager watches the collected abnormal person video through the mobile phone and controls the alarm to continuously alarm or stop alarming through the mobile phone; the processed collected video data are stored in a database, and a manager can check the collected video data through a time stamp after login verification is successful.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310820143.6A CN116863399A (en) | 2023-07-06 | 2023-07-06 | Network security monitoring system and method based on artificial intelligence |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310820143.6A CN116863399A (en) | 2023-07-06 | 2023-07-06 | Network security monitoring system and method based on artificial intelligence |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN116863399A true CN116863399A (en) | 2023-10-10 |
Family
ID=88231643
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310820143.6A Pending CN116863399A (en) | 2023-07-06 | 2023-07-06 | Network security monitoring system and method based on artificial intelligence |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN116863399A (en) |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107590936A (en) * | 2017-09-20 | 2018-01-16 | 深圳市鑫汇达机械设计有限公司 | A kind of warehouse warning system based on video monitoring |
| US20190325198A1 (en) * | 2015-09-22 | 2019-10-24 | ImageSleuth, Inc. | Surveillance and monitoring system that employs automated methods and subsystems that identify and characterize face tracks in video |
| CN112528966A (en) * | 2021-02-05 | 2021-03-19 | 华东交通大学 | Intelligent monitoring and identifying method, device and medium for peripheral environment of payee |
| CN115079609A (en) * | 2022-06-27 | 2022-09-20 | 上海托旺数据科技有限公司 | System and method for managing personnel behaviors and equipment working condition behaviors based on multiple scenes of Internet of things |
-
2023
- 2023-07-06 CN CN202310820143.6A patent/CN116863399A/en active Pending
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US20190325198A1 (en) * | 2015-09-22 | 2019-10-24 | ImageSleuth, Inc. | Surveillance and monitoring system that employs automated methods and subsystems that identify and characterize face tracks in video |
| CN107590936A (en) * | 2017-09-20 | 2018-01-16 | 深圳市鑫汇达机械设计有限公司 | A kind of warehouse warning system based on video monitoring |
| CN112528966A (en) * | 2021-02-05 | 2021-03-19 | 华东交通大学 | Intelligent monitoring and identifying method, device and medium for peripheral environment of payee |
| CN115079609A (en) * | 2022-06-27 | 2022-09-20 | 上海托旺数据科技有限公司 | System and method for managing personnel behaviors and equipment working condition behaviors based on multiple scenes of Internet of things |
Non-Patent Citations (4)
| Title |
|---|
| HIDEKI OKITA等: "AI-based Video Analysis Solution for Creating Safe and Secure Society", Retrieved from the Internet <URL:https://www.hitachi.com/rev/archive/2020/r2020_05/pdf/05b06.pdf> * |
| 杨谦等: "视频异常行为识别与分级预警系统", 科学技术与工程, no. 14, 18 May 2015 (2015-05-18) * |
| 王志国等: "监控视频异常检测：综述", 清华大学学报（自然科学版）, vol. 60, no. 6, 31 December 2020 (2020-12-31) * |
| 王志明: "基于时空特征的异常行为检测", 中国优秀硕士学位论文全文数据库 (信息科技辑), no. 5, 15 May 2023 (2023-05-15) * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108053427B (en) | Improved multi-target tracking method, system and device based on KCF and Kalman | |
| CN110419048B (en) | System for identifying defined objects | |
| CN102811343B (en) | Intelligent video monitoring system based on behavior recognition | |
| CN104966304A (en) | Kalman filtering and nonparametric background model-based multi-target detection tracking method | |
| CN106593534A (en) | Intelligent tunnel construction security monitoring system | |
| CN105100724A (en) | Remote and safe intelligent household monitoring method and device based on visual analysis | |
| CN102254394A (en) | Antitheft monitoring method for poles and towers in power transmission line based on video difference analysis | |
| CN110458794B (en) | Quality detection method and device for accessories of rail train | |
| CN117035419B (en) | Intelligent management system and method for enterprise project implementation | |
| CN114359976B (en) | Intelligent security method and device based on person identification | |
| CN112329691A (en) | Monitoring video analysis method and device, electronic equipment and storage medium | |
| CN114140745A (en) | Method, system, device and medium for detecting personnel attributes of construction site | |
| CN104159088A (en) | System and method of remote monitoring of intelligent vehicle | |
| Patino et al. | Abnormal behaviour detection on queue analysis from stereo cameras | |
| KR102263512B1 (en) | IoT integrated intelligent video analysis platform system capable of smart object recognition | |
| CN110378935B (en) | Parabolic identification method based on image semantic information | |
| CN113392706A (en) | Device and method for detecting smoking and using mobile phone behaviors | |
| CN113052125A (en) | Construction site violation image recognition and alarm method | |
| CN112532927A (en) | Intelligent safety management and control system for construction site | |
| CN116863399A (en) | Network security monitoring system and method based on artificial intelligence | |
| CN111753587A (en) | Method and device for detecting falling to ground | |
| CN110287929B (en) | Method, device, equipment and storage medium for determining number of targets in group area | |
| CN115995093A (en) | Safety helmet wearing identification method based on improved YOLOv5 | |
| CN113128414A (en) | Personnel tracking method and device, computer readable storage medium and electronic equipment | |
| JP6968248B1 (en) | Image analysis report system, image analysis report program and image analysis report method |
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 |