CN115758095A - Multi-dimensional characteristic dynamic abnormal integral model based on Markov-like model - Google Patents
Multi-dimensional characteristic dynamic abnormal integral model based on Markov-like model Download PDFInfo
- Publication number
- CN115758095A CN115758095A CN202310031260.4A CN202310031260A CN115758095A CN 115758095 A CN115758095 A CN 115758095A CN 202310031260 A CN202310031260 A CN 202310031260A CN 115758095 A CN115758095 A CN 115758095A
- Authority
- CN
- China
- Prior art keywords
- information
- time
- vehicle
- abnormal
- model
- 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
- 230000002159 abnormal effect Effects 0.000 title claims abstract description 64
- 239000011159 matrix material Substances 0.000 claims abstract description 21
- 230000007704 transition Effects 0.000 claims abstract description 21
- 238000000034 method Methods 0.000 claims abstract description 20
- 230000005856 abnormality Effects 0.000 claims description 12
- 238000005070 sampling Methods 0.000 claims description 8
- 230000010354 integration Effects 0.000 claims description 5
- 238000006073 displacement reaction Methods 0.000 claims description 4
- 210000000056 organ Anatomy 0.000 claims description 4
- 238000007781 pre-processing Methods 0.000 claims description 4
- 230000035484 reaction time Effects 0.000 claims description 4
- 230000002547 anomalous effect Effects 0.000 claims description 2
- 230000003542 behavioural effect Effects 0.000 claims description 2
- 230000014759 maintenance of location Effects 0.000 claims description 2
- 238000012544 monitoring process Methods 0.000 claims description 2
- 239000013589 supplement Substances 0.000 claims description 2
- 230000002123 temporal effect Effects 0.000 claims description 2
- 230000006399 behavior Effects 0.000 abstract description 40
- 238000005516 engineering process Methods 0.000 abstract description 3
- 230000009286 beneficial effect Effects 0.000 description 1
- 230000000052 comparative effect Effects 0.000 description 1
- 230000003631 expected effect Effects 0.000 description 1
- 238000007689 inspection Methods 0.000 description 1
- 239000000463 material Substances 0.000 description 1
- 230000002265 prevention Effects 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/29—Graphical models, e.g. Bayesian networks
- G06F18/295—Markov models or related models, e.g. semi-Markov models; Markov random fields; Networks embedding Markov models
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N7/00—Computing arrangements based on specific mathematical models
- G06N7/01—Probabilistic graphical models, e.g. probabilistic networks
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T10/00—Road transport of goods or passengers
- Y02T10/10—Internal combustion engine [ICE] based vehicles
- Y02T10/40—Engine management systems
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- Evolutionary Computation (AREA)
- Artificial Intelligence (AREA)
- General Engineering & Computer Science (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Mathematical Optimization (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Life Sciences & Earth Sciences (AREA)
- Probability & Statistics with Applications (AREA)
- Algebra (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Evolutionary Biology (AREA)
- Pure & Applied Mathematics (AREA)
- Computing Systems (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
- Image Analysis (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention discloses a multi-dimensional characteristic dynamic abnormal integral model based on a Markov-like model, which is applied to the prediction analysis of illegal behaviors such as a pass-around card and the like and relates to the technical field of behavior analysis and probability theory. The invention comprises the following steps: the method comprises the steps of firstly, acquiring state information (including position, time, speed and human-vehicle state) of vehicles and personnel by using the Internet of things technology through a GPS, a camera, an electronic network police and an unmanned aerial vehicle, starting from four directions of time characteristics, space path characteristics, human-vehicle state and specific behaviors, and establishing an abnormal integral model of passing-around and avoiding a card by combining randomness of the human behaviors. According to the method, on the basis of space and time characteristics, target state information is analyzed, two constraint conditions of 'human-vehicle state' and 'specific behavior' are innovatively introduced, the transition probability matrix of the Markov-like model is dynamically updated according to abnormal characteristics, the pass-around and card-avoiding behaviors can be effectively predicted, and the pass-around and card-avoiding behaviors can be prevented in time.
Description
Technical Field
The invention relates to a multi-dimensional characteristic dynamic abnormal integral model based on a Markov-like model, which is applied to the prediction analysis of illegal behaviors such as a pass-around card and the like and relates to the technical fields of big data, behavior prediction, probability theory and the like.
Background
Illegal behaviors such as passing and avoiding a card Du Zhi are not absolutely necessary, and the traditional prevention method mainly carries out road and border inspection through manpower, needs to consume a large amount of manpower and material resources and has a poor expected effect. With the development of scientific technology, people can easily acquire the behavior track information of a target object through a GPS (global positioning system) and a camera, and perform face recognition and behavior recognition through edge equipment such as the camera, so that an effective clearance and card avoidance behavior prediction model is established under the space-time characteristics by applying the technologies of the modern Internet of things, the computer and the like, and the method has important theoretical significance and practical significance.
Disclosure of Invention
The invention aims to provide a multi-dimensional characteristic dynamic abnormal integral model based on a Markov-like model, which is applied to the prediction analysis of illegal behaviors such as customs clearance and card avoidance, target state information is analyzed on the basis of space and time characteristics, two constraint conditions of 'human-vehicle state' and 'specific behavior' are innovatively introduced, the Markov-like model changes the form of a Markov model transition probability matrix, the customs clearance card behavior can be effectively predicted, and the occurrence of customs clearance card behavior is prevented in time.
The Markov model of the invention is
，
In the case of the current state of the mobile terminal,
in the last state of the operation, the state,
for the state transition probability matrix, after the first state change
Is a constant value.
The Markov-like model of the invention is
，
In the case of the current state of the mobile terminal,
in the last state of the operation, the state,
as a function of time and state, a state transition probability matrix
Is constantly changing.
In order to achieve the above purpose, the method of the invention comprises the following steps:
step S1: and acquiring related target state information of the personnel and the vehicle, wherein the related target state information comprises vehicle basic information, personnel basic information, position information and time information.
Step S2: data preprocessing is carried out on the target state information to obtain the moving track point of the target object
N is the number of current moving track points and the staying point of the target object
M is the current number of the resident points and the position information of the moving track point
Location information of a dwell point
In which
Representing longitude and latitude information, time information of moving track point
Dwell time at dwell Point position
。
And step S3: initializing abnormal values under the conditions of time characteristics, space characteristics, human-vehicle states and specific behavior characteristics of each node
。
And step S4: according to the relevant state information of the target object obtained in the steps S1 and S2, time feature analysis, space feature analysis, human-vehicle state condition analysis and specific behavior feature analysis are carried out, and abnormal values under different features of each node are updated
。
Step S5: initializing a Markov-like model, defining a state space of
According to the abnormal probability value, further dividing the abnormal data into mild abnormality, moderate abnormality and severe abnormality; an initial probability distribution of
Wherein
Respectively representing the initial normal probability and abnormal probability of the state space, and carrying out different values according to different conditions, such as
(ii) a Abnormal values under the time characteristics, the space characteristics, the human-vehicle state conditions and the specific behavior characteristics of each node updated through S4
Computing a state transition probability matrix
。
Step S6: updating outliers of target objects through markov-like models
，
，
Representing the abnormal probability value of the target object at the i track point,
the probability of normality is indicated by the probability of normality,
indicating the probability of an anomaly, based on
The value of (2) is combined with the established abnormal standard, and the possibility of the illegal behavior of avoiding the card by passing the customs at the node i is further judged.
The specific process of S1 is as follows: specific ways of obtaining the relevant target status information of the person and the vehicle include,
s1.1, acquiring the registration information of vehicles and personnel in vehicle administration centers and traffic management departments of public security organs.
S1.2, acquiring main information of a vehicle track through a vehicle GPS in scenes such as roads, streetscapes and the like, and capturing auxiliary information through a camera and a wireless alarm device; the unmanned aerial vehicle is adopted to search in places where people are difficult to reach (for example, complicated road sections, forest regions, riverways and the like have serious hidden dangers).
The specific process of S2 is as follows: the specific steps for processing the target state information are as follows, setting the time gap
The time is as long as the reaction time is short,
the average velocity in
And s denotes the target in the time gap
Displacement of the inner; dividing the target dynamic track into two classes for sampling, firstly, every two classes are separated
Sampling the target state information and extracting the motion point
To obtain the moving track point of the target object
N is the number of current moving track points; in that
Inner and average velocity
Point set of
Defined as the resident points of the target object, m is the number of the currently determined resident points,
for a customized speed standard, e.g. set to 10km/h, i.e. speed less than
The track points are regarded as residence points; obtaining the position information of the mobile track point according to S1.2
Location information of a dwell point
Wherein
Representing longitude and latitude information and acquiring time information of mobile track point
Dwell time of dwell point position
。
The specific process of S4 is as follows: performing space characteristic analysis, time characteristic analysis, human-vehicle state condition analysis and specific behavior characteristic analysis on the related state information of the target object, and updating abnormal values under each characteristic value
The method comprises the following specific steps: moving track point of target object obtained according to S2
If the target track is close to the sensitive zone, increasing the abnormal value of the spatial characteristic
(ii) a The resident point of the target object is obtained according to the S2
Information that if the target object stays many times at a border point, an unmanned area, or a remote area, the abnormal value of the spatial feature is increased
And specific behavioral outliers
(ii) a The time information of the moving track points obtained according to the S2 is
If the time of going out is in the morning or at night, the abnormal index of the time characteristic according to different time
Updating is carried out; the residence time of the residence point position obtained according to the S2
If the retention time exceeds a certain value, the time characteristic abnormality index is calculated
Updating is carried out; the states of the vehicle and the person are acquired through the vehicle and the mobile phone GPS, information supplement is carried out by electronic monitoring equipment and the like, and the abnormal value of the state of the vehicle and the person is obtained if the vehicle and the person are separated or the driving vehicle and the registered vehicle are found
And (6) updating. Some specific update principles are shown in attached table 1.
The state transition probability matrix of the S5
，
,
Is the outlier of the jth anomalous feature,
is the dominance weight of the anomaly characteristic,
。
the specific process of S6 is as follows: updating abnormal values of target objects by Markov-like models, i.e.
，
Representing the abnormal probability distribution of the target object at the i track point,
indicating probability of normality，
Indicating the probability of an anomaly, based on
The value of (2) is combined with the established abnormal standard, and the possibility of illegal behaviors such as clearance and card avoidance and the like at the node i is further judged; unlike the Markov model that the state transition probability matrix is invariant, the present invention (Markov-like model) updates the transition probability matrix by the spatial characteristics, temporal characteristics, man-vehicle state conditions, and specific behavior characteristics calculated from the target state information of each node, while retaining the characteristic information of the previous position node.
The beneficial effects of the invention are:
the invention provides a multi-dimensional characteristic dynamic abnormal integral model based on a Markov-like model, which is applied to the prediction analysis of illegal behaviors such as a pass-around card and the like; according to the method, on the basis of space and time characteristics, target state information is analyzed, two constraint conditions of 'human-vehicle state' and 'specific behavior' are innovatively introduced, the Markov-like model improves the form of a Markov model transition probability matrix, the pass-by and card-avoiding behaviors can be effectively predicted, and the pass-by and card-avoiding behaviors can be prevented in time.
Drawings
FIG. 1 is a general flow chart of the present invention;
Detailed Description
Table 1 is a partial abnormal update case;
table 2 shows the abnormal state of different abnormal values.
The invention will be further described with reference to the drawings and the detailed description, but the scope of the invention is not limited to the scope.
Example 1
Step S1: acquiring related target state information of personnel and vehicles, wherein the related target state information comprises vehicle basic information, personnel basic information and position information; the specific mode comprises the following steps:
s1.1, acquiring the registration information of vehicles and personnel in vehicle administration centers and traffic management departments of public security organs.
S1.2, acquiring main information of a vehicle track through a vehicle GPS in scenes such as roads, streetscapes and the like, and capturing auxiliary information through a camera and a wireless alarm; the unmanned aerial vehicle is adopted to search in places with serious hidden dangers, such as complex road sections, forest regions, riverways and the like which are difficult to reach by people.
Step S2: the specific steps of carrying out data preprocessing on the target state information and carrying out data processing on the target state information are as follows, and setting a time gap
The time is as long as the reaction time is short,
average velocity of
And s denotes the target in the time gap
Displacement of the inner; dividing the target dynamic track into two classes for sampling, firstly, every two classes are separated
Sampling the target state information and extracting the motion point
To obtain the moving track point of the target object
N is the number of current moving track points; in that
Inner and average velocity
Point set of (2)
Defined as the dwell point of the target object, m is the number of the currently determined dwell points,
the speed standard is customized and is set to be 5km/h, namely the speed is less than
The track points are regarded as residence points; obtaining the position information of the mobile track point according to S1.2
Location information of a dwell point
Wherein
Representing longitude and latitude information and acquiring time information of a mobile track point
Dwell time of dwell point position
。
And step S3: initializing abnormal values under the conditions of time characteristics, space characteristics, human-vehicle states and specific behavior characteristics of each node
。
And step S4: and performing time characteristic analysis, space characteristic analysis, human-vehicle state condition analysis and specific behavior characteristic analysis according to the relevant state information of the target object acquired in the steps S1 and S2. If the target object is analyzed at the M node to obtain that the target object passes through the border key intersection for multiple times, updating the abnormal value of the specific behavior characteristic, namely
And if there is no other abnormal condition, the abnormal value at the M node is
. Then, the target object is analyzed at the N nodes to be found from 1 point to 4 points in the morning, and the driving track is to go to the forest areaIf the abnormal value of the N node is updated to be in an unmanned area such as a river channel and stays in the unmanned area
Table 1: partial abnormal update condition (can be continuously expanded according to actual conditions)
Step S5: initializing a Markov-like model, defining a state space of
And abnormal values under the spatial characteristics, the time characteristics, the human-vehicle state conditions and the specific behavior characteristics of each node updated through S4
Calculating a state transition probability matrix p; i.e. for N nodes, outliers
Then, then
0.15+0.25 + 2+0.2=0.85 due to spatial characteristics
Reaches the maximum, so its corresponding weight is increased to 2,
i.e. by
。
Step S6: updating abnormal values of target objects by Markov-like models, i.e.
，
Indicating that the target object is on the i trackThe abnormal probability distribution of the trace points,
the probability of normality is indicated by the probability of normality,
the probability of representing abnormity is not changed unlike the Markov model state transition probability matrix, the model calculates the space characteristic, the time characteristic, the man-vehicle state condition and the specific behavior characteristic according to the target state information of each node to update the transition probability matrix P, and simultaneously, the characteristic information of the previous position node is also reserved, and the probability is arranged at the N-1 node
Then is obtained by
To obtain
According to
The value of (2) is combined with the established abnormal standard, so that the serious abnormality of the target object can be obtained, and illegal behaviors such as clearance and card avoidance are possibly caused.
Table 2: abnormal state condition of different abnormal values
Comparative examples
In contrast, the specific process of the multi-dimensional characteristic dynamic anomaly integration model using the markov model is the same as that of embodiment 1, except that the conventional markov model is used in step 6, and the specific process is as follows:
step S1: acquiring related target state information of personnel and vehicles, wherein the related target state information comprises vehicle basic information, personnel basic information and position information; the concrete method comprises the following steps:
s1.1, acquiring the registration information of vehicles and personnel in vehicle administration centers and traffic management departments of public security organs.
S1.2, acquiring main information of a vehicle track through a vehicle GPS in scenes such as roads, streetscapes and the like, and capturing auxiliary information through a camera and a wireless alarm; the unmanned aerial vehicle is adopted to search in places with serious hidden dangers, such as complex road sections, forest regions, riverways and the like which are difficult to reach by people.
Step S2: the specific steps of carrying out data preprocessing on the target state information and carrying out data processing on the target state information are as follows, and setting a time gap
The time is as long as the reaction time is short,
average velocity of
And s denotes the target in the time gap
Displacement of the inner; dividing the target dynamic track into two classes for sampling, firstly, every two classes are separated
Sampling the target state information and extracting the motion point
To obtain the moving track point of the target object
N is the number of the current moving track points; in that
Inner, average velocity
Point set of
Defined as the resident points of the target object, m is the number of the currently determined resident points,
the speed standard is customized and is set to be 5km/h, namely the speed is less than
The track points are regarded as residence points; obtaining the position information of the mobile track point according to S1.2
Location information of a dwell point
Wherein
Representing longitude and latitude information and acquiring time information of mobile track point
Dwell time of dwell point position
。
And step S3: initializing abnormal values under the conditions of time characteristics, space characteristics, human-vehicle states and specific behavior characteristics of each node
。
And step S4: and performing time characteristic analysis, space characteristic analysis, human-vehicle state condition analysis and specific behavior characteristic analysis according to the relevant state information of the target object acquired in the steps S1 and S2. Assuming that the time characteristic, the space characteristic, the human-vehicle state condition and the specific behavior characteristic of the target object at the first node of the track sequence are all analyzed normally, the abnormal value of the first position point is still the abnormal value after being updated
。
Step S5: initializing the Markov model, defining a state space of
Respective node null updated by S4Inter-characteristic, time characteristic, human-vehicle state condition, abnormal value under specific behavior characteristic
Calculating a state transition probability matrix p; i.e. for node 1, outlier
Then, then
，
I.e. by
。
Step S6: updating the outliers of the target object by means of Markov models, i.e.
，
Representing the abnormal probability distribution of the target object at the N track points,
the probability of normality is indicated by the probability of normality,
representing the probability of abnormality, setting the probability distribution of abnormality in the initial state
Then is obtained by
To obtain
According to
The value of (2) is combined with the established abnormal standard, so that the target object is normal at the node 1.
Since the state matrix of the Markov model is constant during the moving process of the target, the transition probability matrix of the node 1 will be the transition probability matrix of the whole process, i.e. the transition probability matrix of the whole process
From
Or is obtained
The node N will always be determined to be normal.
However, in practical situations, the target object is at the node N, starts from 0-4 in the morning, the driving track is to a sensitive area such as a forest area, a river channel and the like, and stops in an unmanned area, and the target is very abnormal at this time, which means that the abnormal condition of the subsequent node cannot be judged by using the abnormal value of the initial node.
The node N is calculated to be the severe abnormity through the Markov model provided by the invention, so that the method is more accurate and reasonable.
Compared with the prior art, the multi-dimensional characteristic dynamic abnormal integral model based on the Markov-like model dynamically calculates the state transition probability of the current point according to the abnormal value of each track point, so that the problem that the state transition probability matrix of the traditional Markov model cannot be changed can be solved, and meanwhile, two constraint conditions of 'human-vehicle state' and 'specific behavior' are introduced on the basis of the space-time characteristic, so that the illegal behavior of avoiding the passing-around customs can be more accurately predicted and judged.
While the present invention has been described in detail with reference to the embodiments shown in the drawings, the present invention is not limited to the embodiments, and various changes can be made without departing from the spirit of the present invention within the knowledge of those skilled in the art.
Claims (5)
1. A multi-dimensional characteristic dynamic abnormal integral model based on a Markov-like model is characterized in that target state information is analyzed on the basis of space and time characteristics, two constraint conditions of 'man-vehicle state' and 'specific behavior' are introduced simultaneously, and the form of a transition probability matrix of the original Markov model is improved through characteristic abnormal values, and the method comprises the following specific steps:
step S1: acquiring related target state information of personnel and vehicles, wherein the related target state information comprises vehicle basic information, personnel basic information, position information and time information;
step S2: data preprocessing is carried out on the target state information to obtain the moving track point of the target object
N is the number of current moving track points and the staying point of the target object
M is the current number of the resident points and the position information of the moving track point
Location information of a dwell point
Wherein
Representing latitude and longitude information, time information of moving track point
Dwell time of dwell point position
；
And step S3: initializing abnormal values under the conditions of time characteristics, space characteristics, human-vehicle states and specific behavior characteristics of each node
；
And step S4: according to the relevant state information of the target object obtained in the steps S1 and S2, time feature analysis, space feature analysis, human-vehicle state condition analysis and specific behavior feature analysis are carried out, and abnormal values under different features of each node are updated
；
Step S5: initializing a Markov-like model, defining a state space of
According to the abnormal probability value, further dividing the abnormal data into mild abnormality, moderate abnormality and severe abnormality; an initial probability distribution of
Wherein
Respectively representing the probability of initial normality and the probability of abnormality of a state space; abnormal values under the time characteristics, the space characteristics, the human-vehicle state conditions and the specific behavior characteristics of each node updated through S4
Computing a state transition probability matrix
；
Step S6: updating abnormal values of target objects by Markov-like models, i.e.
，
，
Representing the abnormal probability distribution of the target object at the i track point,
the probability of being normal is indicated and,
indicating the probability of an anomaly based on
The value of (2) is combined with the established abnormal standard, and the possibility of the illegal behavior of avoiding the card by passing the customs at the node i is further judged.
2. The markov model-like based multi-dimensional feature dynamic anomaly integration model of claim 1, wherein: the specific mode of S1 for acquiring the related target state information of the personnel and the vehicle is as follows:
s1.1, acquiring registration information of vehicles and personnel in a vehicle administration post and a traffic management department of a public security organ;
s1.2, acquiring main information of a vehicle track through a vehicle GPS in scenes such as roads, streetscapes and the like, capturing auxiliary information through a camera and a wireless alarm device, and searching in places where people are difficult to reach by adopting an unmanned aerial vehicle.
3. The markov model-like based multi-dimensional feature dynamic anomaly integration model of claim 2, wherein: the specific steps of the S2 for carrying out data processing on the target state information are as follows:
setting a time gap
The time is as long as the reaction time is short,
average velocity of
And s represents the target in the time slot
Displacement of the inner; dividing the target dynamic track into two classes for sampling, firstly, every two classes are separated
Sampling the target state information and extracting the motion point
To obtain the moving track point of the target object
N is the number of current moving track points;
in that
Inner and average velocity
Point set of
Defined as the resident points of the target object, m is the number of the currently determined resident points,
the speed is a self-defined speed standard;
obtaining the position information of the mobile track point according to S1.2
Location information of a dwell point
Wherein
Representing longitude and latitude information and acquiring time information of mobile track point
Dwell time at dwell Point position
。
4. The Markov-model-like multi-dimensional feature dynamic anomaly integration model according to claim 1, wherein S4 performs spatial feature analysis, temporal feature analysis, analysis of human-vehicle state and specific behavior feature analysis on the state information related to the target object, and updates the anomaly values at the respective feature values
，
The method comprises the following specific steps:
moving track point of target object obtained according to S2
If the target track is close to the sensitive zone, increasing the abnormal value of the spatial characteristic
；
The resident point of the target object is obtained according to the S2
Information if the target object is in a border point, unmanned area, remote areaMultiple stops increase the abnormal value of the spatial feature
And specific behavioral outliers
；
The time information of the moving track points obtained according to the S2 is
If the time of going out is in the morning or at night, the abnormal index of the time characteristic according to different time
Updating is carried out;
the residence time of the residence point position obtained according to the S2
If the retention time exceeds a certain value, the time characteristic abnormality index is calculated
Updating is carried out;
the states of the vehicle and the person are acquired through the vehicle and the mobile phone GPS, information supplement is carried out by electronic monitoring equipment and the like, and the abnormal value of the state of the vehicle and the person is obtained if the vehicle and the person are separated or the driving vehicle and the registered vehicle are found
And (6) updating.
5. The Markov model-like multi-dimensional feature dynamic anomaly integration model of claim 1, wherein the state transition probability matrix of step S5
，
,
Is the outlier of the jth anomalous feature,
is the dominance weight of the anomaly characteristic,
。
Priority Applications (2)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310031260.4A CN115758095A (en) | 2023-01-10 | 2023-01-10 | Multi-dimensional characteristic dynamic abnormal integral model based on Markov-like model |
| CN202310543920.7A CN116502055B (en) | 2023-01-10 | 2023-05-15 | Multi-dimensional characteristic dynamic abnormal integral model based on quasi-Markov model |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202310031260.4A CN115758095A (en) | 2023-01-10 | 2023-01-10 | Multi-dimensional characteristic dynamic abnormal integral model based on Markov-like model |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN115758095A true CN115758095A (en) | 2023-03-07 |
Family
ID=85348838
Family Applications (2)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310031260.4A Pending CN115758095A (en) | 2023-01-10 | 2023-01-10 | Multi-dimensional characteristic dynamic abnormal integral model based on Markov-like model |
| CN202310543920.7A Active CN116502055B (en) | 2023-01-10 | 2023-05-15 | Multi-dimensional characteristic dynamic abnormal integral model based on quasi-Markov model |
Family Applications After (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202310543920.7A Active CN116502055B (en) | 2023-01-10 | 2023-05-15 | Multi-dimensional characteristic dynamic abnormal integral model based on quasi-Markov model |
Country Status (1)
| Country | Link |
|---|---|
| CN (2) | CN115758095A (en) |
Cited By (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116911511A (en) * | 2023-09-14 | 2023-10-20 | 中建三局信息科技有限公司 | Commercial concrete transportation vehicle real-time management method, device, equipment and storage medium |
Family Cites Families (20)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US4783804A (en) * | 1985-03-21 | 1988-11-08 | American Telephone And Telegraph Company, At&T Bell Laboratories | Hidden Markov model speech recognition arrangement |
| US6212510B1 (en) * | 1998-01-30 | 2001-04-03 | Mitsubishi Electric Research Laboratories, Inc. | Method for minimizing entropy in hidden Markov models of physical signals |
| CN101795460B (en) * | 2009-12-23 | 2013-01-30 | 大连理工大学 | Markov mobility model suitable for mobile Ad Hoc network in obstacle environment |
| US9110817B2 (en) * | 2011-03-24 | 2015-08-18 | Sony Corporation | Method for creating a markov process that generates sequences |
| CN104881711B (en) * | 2015-05-18 | 2018-08-07 | 中国矿业大学 | Underground early warning mechanism method based on miner's behavioural analysis |
| CN105956682A (en) * | 2016-04-19 | 2016-09-21 | 上海电力学院 | Short-period electricity price prediction method based on BP neural network and Markov chain |
| CN106447137A (en) * | 2016-11-18 | 2017-02-22 | 广东省科技基础条件平台中心 | Traffic passenger flow forecasting method based on information fusion and Markov model |
| CN107742193B (en) * | 2017-11-28 | 2019-08-27 | 江苏大学 | A kind of driving Risk Forecast Method based on time-varying state transition probability Markov chain |
| CN108665069B (en) * | 2018-04-24 | 2022-04-15 | 东南大学 | Sudden event triggering mechanism for unmanned vehicle training simulation |
| CN111368290B (en) * | 2018-12-26 | 2023-06-09 | 中兴通讯股份有限公司 | Data anomaly detection method and device and terminal equipment |
| CN109902744A (en) * | 2019-02-28 | 2019-06-18 | 成都新希望金融信息有限公司 | A method of it is modified based on exceptional value of the Markov transition matrix to matrix |
| CN111913859B (en) * | 2020-07-13 | 2023-11-14 | 北京天空卫士网络安全技术有限公司 | Abnormal behavior detection method and device |
| CN111859709B (en) * | 2020-07-31 | 2022-10-18 | 河北地质大学 | Geologic statistics simulation method and device for aquifer structure variation transition probability |
| CN111915104A (en) * | 2020-08-28 | 2020-11-10 | 山东省国土测绘院 | Method and device for predicting outgoing position |
| CN112070073B (en) * | 2020-11-12 | 2021-02-02 | 北京中恒利华石油技术研究所 | Logging curve abnormity discrimination method based on Markov chain transition probability matrix eigenvalue classification and support vector machine |
| CN113505935B (en) * | 2021-07-26 | 2022-06-28 | 上海东方低碳科技产业股份有限公司 | Electric power abnormal fluctuation detection and prediction calculation method based on integrated algorithm |
| CN114676743B (en) * | 2021-12-09 | 2024-04-26 | 上海无线电设备研究所 | Low-speed small target track threat identification method based on hidden Markov model |
| CN114510965A (en) * | 2022-01-12 | 2022-05-17 | 硕橙（厦门）科技有限公司 | Abnormal sound detection method, device, equipment and medium |
| CN115019238A (en) * | 2022-07-04 | 2022-09-06 | 南京航空航天大学 | Group target dynamic behavior identification method based on hidden Markov model |
| CN115293639A (en) * | 2022-08-26 | 2022-11-04 | 中国航天科工集团八五一一研究所 | Battlefield situation studying and judging method based on hidden Markov model |
-
2023
- 2023-01-10 CN CN202310031260.4A patent/CN115758095A/en active Pending
- 2023-05-15 CN CN202310543920.7A patent/CN116502055B/en active Active
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN116911511A (en) * | 2023-09-14 | 2023-10-20 | 中建三局信息科技有限公司 | Commercial concrete transportation vehicle real-time management method, device, equipment and storage medium |
| CN116911511B (en) * | 2023-09-14 | 2023-12-12 | 中建三局信息科技有限公司 | Commercial concrete transportation vehicle real-time management method, device, equipment and storage medium |
Also Published As
| Publication number | Publication date |
|---|---|
| CN116502055B (en) | 2024-05-03 |
| CN116502055A (en) | 2023-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10417816B2 (en) | System and method for digital environment reconstruction | |
| US11449727B2 (en) | Method, storage medium and electronic device for detecting vehicle crashes | |
| Shah et al. | Automated visual surveillance in realistic scenarios | |
| Zhang et al. | Prediction of pedestrian-vehicle conflicts at signalized intersections based on long short-term memory neural network | |
| Sun et al. | Vehicle reidentification using multidetector fusion | |
| Mahaur et al. | Road object detection: a comparative study of deep learning-based algorithms | |
| Yu et al. | Trajectory outlier detection approach based on common slices sub-sequence | |
| Wang et al. | A hidden Markov model for urban-scale traffic estimation using floating car data | |
| CN110689043A (en) | Vehicle fine granularity identification method and device based on multiple attention mechanism | |
| Méneroux et al. | Traffic signal detection from in-vehicle GPS speed profiles using functional data analysis and machine learning | |
| Huo et al. | Short-term estimation and prediction of pedestrian density in urban hot spots based on mobile phone data | |
| Spannaus et al. | AUTOMATUM DATA: Drone-based highway dataset for the development and validation of automated driving software for research and commercial applications | |
| CN115758095A (en) | Multi-dimensional characteristic dynamic abnormal integral model based on Markov-like model | |
| Dai et al. | Attention based simplified deep residual network for citywide crowd flows prediction | |
| Zhu et al. | Spatio-temporal point processes with attention for traffic congestion event modeling | |
| Khan et al. | Short-term traffic prediction using deep learning long short-term memory: Taxonomy, applications, challenges, and future trends | |
| US20180260401A1 (en) | Distributed video search with edge computing | |
| Wang et al. | Traffic speed estimation based on multi-source GPS data and mixture model | |
| Zhang | Multi-object trajectory extraction based on YOLOv3-DeepSort for pedestrian-vehicle interaction behavior analysis at non-signalized intersections | |
| Ar et al. | A computer vision-based object detection and counting for COVID-19 protocol compliance: a case study of Jakarta | |
| Yankun et al. | A color histogram based large motion trend fusion algorithm for vehicle tracking | |
| Cerqueira et al. | Integrative analysis of traffic and situational context data to support urban mobility planning | |
| CN113689705B (en) | Method and device for detecting red light running of vehicle, computer equipment and storage medium | |
| EP3410362B1 (en) | Method and apparatus for next token prediction based on previously observed tokens | |
| CN104966264A (en) | Security big data based on text serving as transformation basic state of heterogeneous data for fusion processing |
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 | ||
| WD01 | Invention patent application deemed withdrawn after publication | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20230307 |