CN110309953B - Urban security monitoring layout system and method adopting target mobility distribution prediction - Google Patents
Urban security monitoring layout system and method adopting target mobility distribution prediction Download PDFInfo
- Publication number
- CN110309953B CN110309953B CN201910453560.5A CN201910453560A CN110309953B CN 110309953 B CN110309953 B CN 110309953B CN 201910453560 A CN201910453560 A CN 201910453560A CN 110309953 B CN110309953 B CN 110309953B
- Authority
- CN
- China
- Prior art keywords
- target
- individual
- external
- flowing
- flow
- 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.)
- Active
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q10/00—Administration; Management
- G06Q10/04—Forecasting or optimisation specially adapted for administrative or management purposes, e.g. linear programming or "cutting stock problem"
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
- G06Q50/00—Systems or methods specially adapted for specific business sectors, e.g. utilities or tourism
- G06Q50/10—Services
- G06Q50/26—Government or public services
- G06Q50/265—Personal security, identity or safety
Abstract
The invention discloses an urban security monitoring layout system adopting target mobility distribution prediction, which inputs external mobility target characteristic quantity capable of reflecting the degree of influence of external mobility targets in a set area on individual mobility into various individual mobility behavior neural network models, then predicts the corresponding mobility individual quantity distribution generating mobility behaviors on each external mobility target in the set area based on the mobility behavior tendency of each individual output by each neural network model to the corresponding external mobility target, and finally screens out the external mobility targets with the mobility individual quantity reaching the standard, and generates a security monitoring system layout scheme of the corresponding urban area based on the external mobility targets reaching the standard. The system can simulate the mobility of people in a space range more scientifically and more truly and accurately, and the monitoring system is installed between the initial targets with the number reaching the standard and the correlation parties consisting of the external mobile targets, so that the installation layout of the monitoring system is optimized.
Description
Technical Field
The invention relates to the technical field of security monitoring, in particular to an urban security monitoring layout system adopting target mobility distribution prediction and an urban security monitoring layout method adopting target mobility distribution prediction.
Background
The existing crowd-oriented video security monitoring system is widely installed and applied in cities, clear video images can be collected, and technologies such as face recognition and behavior recognition can be matched to perform identity identification and abnormity alarm, so that social security is maintained, crimes are deterred, and life and property safety of people is guaranteed.
Regarding the layout problem of security monitoring systems in cities, economic and material conditions of many cities at present are difficult to realize the complete coverage of the whole space of the city without dead angles and blind areas, so that the security video shooting and collecting effect is maximized when the security monitoring systems are laid, and the security monitoring systems are prevented from being laid too densely, the construction difficulty is increased, the resource waste is caused, and the security monitoring layout optimization problem needs to be considered.
The distribution and point selection of the security monitoring system obviously takes the mobility rule of people in the city range into consideration. Specifically, if a predetermined number (for example, 100 security monitoring systems are expected to be installed) of security monitoring systems are expected to be installed in a defined space range (for example, in an administrative division range of an urban area), the security monitoring systems should be arranged in a place with high crowd mobility in the space range, so that an optimal collection effect can be achieved.
For the mobility of people in different places in a space range, the evaluation is mainly carried out by the design experience of a planning construction party at present, and a scientific quantitative analysis method is lacked. Particularly, the mobility of people in each place in a spatial range is not constant, many cities in China are in a high-speed development period at present, and the mobility of people is continuously changed along with the construction of a new path and a new station of public transportation, the improvement of commercial service matching and the adjustment of distribution of an industrial park and a residential community, so that the predictive analysis of the mobility of people in the spatial range of the city needs to be enhanced, and an optimized security monitoring system layout scheme is designed.
Disclosure of Invention
Based on this, in order to keep pace with the fast updating speed of urban facilities, the flow distribution of the crowd in the space range is predicted more accurately and truly by a more scientific method, and then an optimized layout scheme of the security monitoring system is designed, the application discloses the following technical scheme.
As a first aspect of the present invention, the present invention discloses an urban security monitoring layout system employing target fluidity distribution prediction, comprising:
the tendency calculation module is used for inputting the characteristic quantity of the external flowing target capable of reflecting the influence degree of the external flowing target in the set area on the flow of the individual into various individual flowing behavior neural network models to obtain the flowing behavior tendency of each individual to the corresponding external flowing target;
the flow predicting module is used for predicting the corresponding flow individual quantity distribution of flow behavior generated on each external flow target in the set area based on the flow behavior tendency degree;
the scheme generation module is used for screening out external mobile targets with the mobile individual number reaching the standard and generating a security monitoring system layout scheme of the corresponding urban area based on the external mobile targets reaching the standard;
the various neural network models can respectively correspond to various individuals with different personal conditions, and the layout scheme can at least determine the installation area of the security monitoring system.
In one possible embodiment, the system further comprises:
the flow target screening module is used for screening out urban facilities which meet any one of target types and meet the scale requirement from all the urban facilities in the set area as the external flow targets; wherein the content of the first and second substances,
the target type includes at least one of: transportation facilities, medical facilities, residential facilities, commercial and service facilities, educational facilities.
In one possible embodiment, the system further comprises:
and the characteristic establishing module is used for selecting an initial target from urban facilities as an individual flowing initial position and establishing the external flowing target characteristic quantity of the external flowing target in the set area to the individual.
In one possible embodiment, the external flow target characteristic quantity is expressed as:
wherein i ═ 1, k]K is the target number of external flows within the set area,
for the i-th influence factor of said external flow target on the individual, diIs the distance of the path between the ith external flow target and the individual, the impact factor and diIn inverse proportion, βiAn influence magnitude for the ith external flow target itself, the influence magnitude being proportional to the ability of the external flow target to perform the corresponding function.
In one possible embodiment, the system further comprises:
the individual dividing module is used for dividing the types of the individuals according to the individual conditions of the individuals;
the model establishing module is used for establishing the individual flow behavior neural network model according to the individual type divided by the individual dividing module as a model type so that the various neural network models can respectively correspond to various individuals with different individual conditions;
wherein the personal condition comprises at least one of: age, work status, travel mode, income level, family status, health status, consumption demand level.
In one possible embodiment, the system further comprises a model training module for training the neural network model, wherein the model training module comprises:
the sample determining unit is used for determining a plurality of individuals with different personal conditions as individual samples;
a data acquisition unit configured to acquire the actual flow behavior of the individual sample determined by the sample determination unit and calculate the external flow target feature quantity corresponding to the actual flow behavior;
and the model training unit is used for inputting the external flowing target characteristic quantity calculated by the data acquisition unit into the neural network model, and adjusting the neuron parameters of the neural network model based on the matching degree between the flowing behavior tendency degree output by the neural network model and the actual flowing behavior of the corresponding sample individual acquired by the data acquisition unit until the matching degree meets the requirement.
In one possible embodiment, the individual ambulatory behavior neural network model employs a BP neural network model.
In one possible embodiment, the scenario generation module screens the external flow target meeting the standard using an individual number threshold or an individual number ranking or an individual number proportion threshold.
In a possible implementation manner, the layout scheme further includes an installation number of the security monitoring system, wherein the installation number is determined at least according to the corresponding number of mobile individuals or the ratio of the corresponding number of mobile individuals of each of the qualified external mobile targets.
In one possible embodiment, the plan generation module determines the installation area in the layout plan at least in conjunction with map information of the set area.
In one possible embodiment, the system further comprises:
the prediction data integration module is used for integrating the corresponding flow individual quantity distribution generated by the flow prediction module under a plurality of different scenes to obtain integral quantity distribution; wherein the content of the first and second substances,
factors that distinguish whether the scenarios are different include: individual flow direction, and/or extent of the defined region.
As a second aspect of the present invention, the present invention discloses an urban security monitoring layout method using target fluidity distribution prediction, comprising:
inputting external flowing target characteristic quantity capable of reflecting the influence degree of the external flowing target in the set area on the flow of the individual into various individual flowing behavior neural network models;
predicting the corresponding flowing individual quantity distribution of the flowing behavior generated on each external flowing target in the set area based on the flowing behavior tendency degree of each individual output by each neural network model to the corresponding external flowing target;
screening out external mobile targets with the mobile individual number reaching the standard, and generating a security monitoring system layout scheme of a corresponding urban area based on the external mobile targets reaching the standard;
the various neural network models can respectively correspond to various individuals with different personal conditions, and the layout scheme can at least determine the installation area of the security monitoring system.
In one possible embodiment, the method further comprises:
screening out urban facilities which meet any one of target types and meet the scale requirement from all the urban facilities in the set area as the external flowing target; wherein the content of the first and second substances,
the target type includes at least one of: transportation facilities, medical facilities, residential facilities, commercial and service facilities, educational facilities.
In one possible embodiment, the method further comprises:
and selecting a starting target from urban facilities as an individual flowing starting position, and establishing the external flowing target characteristic quantity of the external flowing target in the set area to the individual.
In one possible embodiment, the external flow target characteristic quantity is expressed as:
In one possible embodiment, the method further comprises:
dividing the types of the individuals according to the individual conditions of the individuals;
establishing the individual flowing behavior neural network model according to the divided individual types as model types, so that the various neural network models can respectively correspond to various individuals with different individual conditions;
wherein the personal condition comprises at least one of: age, work status, travel mode, income level, family status, health status, consumption demand level.
In one possible embodiment, the method further comprises:
training the neural network model, wherein the training the neural network model comprises:
determining a plurality of individuals with different personal conditions as individual samples;
acquiring the actual flow behavior of the individual sample determined by the sample determination unit, and calculating the external flow target characteristic quantity corresponding to the actual flow behavior;
and inputting the external flowing target characteristic quantity calculated by the data acquisition unit into the neural network model, and adjusting neuron parameters of the neural network model based on the matching degree between the flowing behavior tendency degree output by the neural network model and the actual flowing behavior of the corresponding sample individual acquired by the data acquisition unit until the matching degree meets the requirement.
In one possible embodiment, the individual ambulatory behavior neural network model employs a BP neural network model.
In one possible embodiment, the external flow target reaching the standard is screened by using an individual number threshold or an individual number ranking or an individual number ratio threshold.
In a possible implementation manner, the layout scheme further includes an installation number of the security monitoring method, where the installation number is determined at least according to the corresponding number of mobile individuals or the ratio of the corresponding number of mobile individuals of each of the qualified external mobile targets.
In one possible embodiment, the installation area in the layout plan is determined in combination with map information of the set area.
In one possible embodiment, the method further comprises:
integrating the number distribution of the corresponding mobile individuals generated under a plurality of different scenes to obtain integral number distribution, and screening the external mobile target reaching the standard based on the integral number distribution; wherein the content of the first and second substances,
factors that distinguish whether the scenarios are different include: individual flow direction, and/or extent of the defined region.
According to the embodiment of the urban security monitoring layout system and the method, the individual flowing behavior tendency degree in the region is calculated through the neural network model, the individual quantity flowing to each external flowing target is predicted according to the flowing behavior tendency degree, the mobility of people in the space range can be simulated scientifically and truly and accurately, the monitoring system is installed between the initial target with the individual quantity up to the standard and the associated party consisting of the external flowing targets, the urban element high variability caused by high-speed urban development can be adapted, the installation layout of the monitoring system is optimized, the monitoring effect can be maximized, and the cost can be saved.
Drawings
The embodiments described below with reference to the drawings are exemplary and intended to be used for explaining and illustrating the present invention and should not be construed as limiting the scope of the present invention.
Fig. 1 is a block diagram of a first embodiment of a city security monitoring layout system disclosed in the present application.
FIG. 2 is a graph of input-output relationships of an individual flow behavior neural network model.
Fig. 3 is a schematic flow chart of a first embodiment of a city security monitoring layout method disclosed in the present application.
Detailed Description
In order to make the implementation objects, technical solutions and advantages of the present invention clearer, the technical solutions in the embodiments of the present invention will be described in more detail below with reference to the accompanying drawings in the embodiments of the present invention.
The first embodiment of the urban security monitoring layout system adopting target liquidity distribution prediction disclosed by the invention is described in detail below with reference to fig. 1-2.
As shown in fig. 1-2, the city security monitoring layout system disclosed in this embodiment mainly includes: the system comprises a tendency calculation module, a flow prediction module and a scheme generation module.
And the tendency degree operation module is used for inputting the characteristic quantity of the external flowing target into the neural network models of the flowing behaviors of various individuals to obtain the flowing behavior tendency degree of each individual to the corresponding external flowing target. Before the layout of city security monitoring, a set area needs to be preset as a city space range for installing a video monitoring system. The application understands the mobility of the population in the urban space range as: each individual character can form flowing behaviors in a certain urban space range under the stimulation of the internal and external flowing targets, and the flowing behaviors of a large number of individuals are gathered to form macroscopic crowd mobility.
The external floating targets (also referred to as floating targets) are city elements, city units, and city facilities that can have a certain influence on the behavior of individual persons, such as traveling, and the external floating target feature amount can reflect the degree of influence of each external floating target in the set area on the individual persons in terms of floating, and is generally obtained from data such as the type and the number of each external floating target.
The floating targets typically arise in connection with the construction of urban infrastructure and public service facilities associated with urban planning, urban construction, and urban operation. For example, a subway station and an airport are opened in a certain administrative area X, and a technology park is also opened in the administrative area X, so that the subway station, the airport and the technology park are part of flowing targets in the area, the flowing targets have the capability of periodically or even all-weather attracting non-negligible or even huge individual personnel, and the personnel in the set area can go (i.e. flow) to a certain flowing target according to the self needs to realize the self purpose, such as going to the airport for business or traveling, going to the subway station and the technology park for working, and the like.
The individual flowing behavior is the behavior of flowing in a certain urban space range due to the change of the flowing target in the individual, namely, the internal cause determines the response of the individual to the flowing target stimulus. However, the influence of the movement target on individuals with different internal causes is different, and when the urban ecological items such as traffic conditions, employment projects and business environments and the like in a certain urban area are different, the personnel movement behaviors in the urban area are correspondingly different. Taking the traffic conditions in the urban area as an example, for example, a certain type of individual prefers to sit on a subway for traveling, if a subway station is located in the set area, the mobility of the type of individual is greatly influenced, and a bus station located in the set area has no significant influence on the mobility of the type of individual. On the contrary, another type of individuals may prefer to sit for a bus trip, the bus station in the set area has a great influence on the mobility of the type of individuals, and the subway station in the set area has no significant influence on the mobility of the type of individuals. Therefore, when the moving targets of the subway station and the bus station are faced, the moving behaviors of the former type of individuals are more likely to be inclined, and the moving behaviors of the latter type of individuals are more likely to be not inclined.
Further, taking the business conditions of the urban area as an example, for example, a certain class of individuals (usually, individuals with abundant assets) may flow to the mobile targets meeting the demands of high-quality living pursuits and consumption due to the existence of high-grade food blocks, shopping centers, etc., to go to and go out of the high-consumption places; another class of individuals, often less-valued individuals, may move to these mobile goals, which meet their cost-effective pursuit of life and consumption needs, and to these low-consumption locations, due to the presence of flat-rate market and mobile pavements. The above-mentioned life-pursuit and consumption needs are a reflection of the influence of the individual's internal factors on the individual's mobility.
As can be seen from the above, the flow behavior selected by different individuals in the face of the same flow objective is different because of the different internal causes of the individuals. Therefore, when the system is used for flow distribution prediction, a plurality of different types of individuals need to be considered, so that more types of individuals can be taken into a sample range when the flow behavior of the individuals is predicted, and finally, the accuracy of the flow distribution prediction result is increased.
Correspondingly, because the internal causes of different individuals are different, the flow behavior changes are also various and are difficult to be represented by a single individual behavior model, so that the system is pre-established with a plurality of types of individual flow behavior neural network models (hereinafter also referred to as neural network models), each type of neural network model respectively corresponds to a plurality of individuals with different personal conditions, each type of neural network model corresponds to a certain type of individual, and reflects the internal causes of the certain type of individual, such as the income level and the like.
After various individual flowing behavior neural network models capable of predicting flowing behaviors of various different individuals and flowing target characteristic quantities in a preset region are acquired, the characteristic quantities are input into the various neural network models, and output of the various neural network models is obtained, namely flowing behavior tendency degrees of the different individuals to the current flowing targets participating in calculation are obtained. The flowing behavior tendency degree represents the selection attitude of an individual for the flowing target, and the higher the tendency degree, the more positive the selection attitude of the flowing individual for the flowing target is, namely the higher the priority of the flowing target is, the higher the flowing individual has the flowing probability to the flowing target at the time of selection.
Therefore, the individual flow behavior neural network model is a multi-input multi-output model, and a multilayer neural network model can be adopted. The input of the neural network model is a multi-dimensional vector 'external flowing target characteristic quantity', wherein the number of dimensions is related to the number of flowing targets, and the output of the neural network model is an m-dimensional vector consisting of flowing behavior tendency degrees of individuals to the 1 st to the mth flowing targets.
Specifically, assuming that the layout of a security monitoring system for D1 administrative districts in D city is required currently, the target liquidity distribution prediction needs to be performed on D1 administrative districts. As shown in fig. 2, assuming that m is 50 floating targets currently in the d1 administrative district, then a certain number and types of the "individual floating behavioral neural network models" are put in, wherein the type, number and putting position of the put-in individuals can be realized based on the result of the sampling survey, so as to make the selected individuals have certain representativeness as possible and match with the real situation of the set area.
For example, according to the sampling survey of individuals in the set area, the individuals are divided into n-24 different types according to the survey result, and a total of 20000 individuals are selected from the d1 administrative area as data samples in a sampling mode, wherein 1000 individuals belong to the 1 st class, 600 individuals belong to the 2 nd class … …, and 300 individuals belong to the 24 th class, and then the inclination degree calculation module puts the corresponding 20000 neural network models into the d1 administrative area and inputs the external flow target characteristic quantity for calculation.
It can be understood that, since it takes a certain time to install and construct the security monitoring system, when the tendency calculation module selects the moving target in the d1 administrative district, in addition to the moving target that already exists currently, the moving target that is in the planning or being constructed may also be selected, such as a subway station that is expected to be officially opened and operated after 3 months, or a shopping mall that is expected to be officially opened after 1 month.
After the operation is completed, the flowing behavior tendency is obtained, for example, the flowing behavior tendency of the class 1 neural network model to the class 1 flowing target is 90, the flowing behavior tendency of the class 1 neural network model to the class 2 flowing target is 50, the flowing behavior tendency of the class 2 neural network model to the class 1 community is 60, and so on, the flowing behavior tendency of 50 × 24 types is obtained, and finally, 50 × 20000 flowing behavior tendency degrees are obtained through statistics.
The flow prediction module is used for predicting the corresponding flow individual quantity distribution of flow behavior generated on each external flow target in the set area based on the flow behavior tendency degree of each body output by each neural network model to the corresponding external flow target.
And after the flowing behavior tendency degrees are obtained, judging whether the magnitude of each flowing behavior tendency degree meets a corresponding flowing behavior threshold value. When the flowing behavior tendency of the ith type of neural network model (i ═ 1, 24) to the jth flowing target (j ═ 1, 50) meets the tendency threshold, it indicates that the individual of the type corresponding to the ith type of neural network model can be predicted to actually generate the behavior flowing to the jth flowing target, and when the flowing behavior tendency does not meet the tendency threshold, it indicates that the corresponding flowing behavior cannot be actually generated. For example, the class 1 neural network model corresponds to ten thousands of local young individuals, and the total number of the local young individuals is 1000, the 9 th floating target is a certain top-grade shopping mall, the floating behavior tendency of the class 1 individual to the 9 th floating target is 85, which is greater than 70 of the tendency threshold, and then the 1000 individuals are predicted to come in and go out (i.e. flow in) the top-grade shopping mall.
By traversing the flowing behavior tendency degrees of the 24 types of neural network models to the 50 flowing targets, the flowing behavior corresponding relation of each individual in the various types of individuals to each external flowing target can be predicted, that is, which type of individual in the individuals corresponding to the 24 types of neural network models can be predicted to generate the flowing behavior to which flowing target, so as to obtain the flowing distribution prediction data of the d1 administrative district. For example, the 20000 individuals are located in a cell, wherein the 1 st, 3 rd, 9 th class (5000 total) of individuals are predicted to flow between the cell and the P subway station (1 st flowing target), the 2 nd, 3 rd, 6 th, 7 th class (9000 total) of individuals are predicted to flow between the cell and the Q shopping mall station (2 nd flowing target), and so on.
The scheme generation module is used for screening out external mobile targets with the mobile individual number reaching the standard according to the mobile individual number predicted by the mobile prediction module, and generating a security monitoring system layout scheme of the corresponding urban area based on the external mobile targets with the mobile individual number reaching the standard.
Individual flow quantities are distributed from the cell to each flow target, and the flow quantities are more or less, so that the flow targets with the flow quantities reaching the standard need to be judged by utilizing a preset standard, the layout of the monitoring system is only carried out aiming at the flow targets reaching the standard, the best use is realized when the number of the monitoring systems is limited, the monitoring effect is maximized, or the achievement of the flow targets is realized when the number of the monitoring systems is rich, and the overall cost is reduced. The standard for judging whether the target reaches the standard is generally based on the predicted number of the mobile individuals, and the mobile targets in the correlation parties with larger numbers are easier to judge as the target which reaches the standard, because the larger the individual flow is, the better the monitoring effect is.
The layout scheme can at least determine the installation area of the security monitoring system. For example, installing a monitoring system somewhere between the starting target and the compliance flow target. The plan generation module determines an installation area in the layout plan at least in conjunction with map information of the set area. For example, map information including navigation information and the like between a starting object and a shopping center determined as a standard-reaching moving object is acquired, then a path which is easier to select, such as a highway, a trunk road, a large crossroad and the like, is selected, and a monitoring system is installed in a proper area of the path, so that a monitoring effect is improved.
The system calculates the individual flowing behavior tendency degree in the region through the neural network model, predicts the individual quantity flowing to each external flowing target according to the individual flowing behavior tendency degree, can simulate the mobility of people in a space range more scientifically and more truly and accurately, and installs the monitoring system between the initial target with the standard individual quantity and the associated party consisting of the external flowing targets, can adapt to the high variability of urban elements caused by the high-speed development of cities, optimizes the installation layout of the monitoring system, can ensure the maximization of the monitoring effect and can save the cost.
For urban facilities such as remote bus stations and cold supermarkets, the flow suction capacity is poor, and the monitoring effect of the monitoring system installed nearby the facilities is small, so in one implementation mode, the system further comprises a flow target screening module, and the flow target screening module is used for screening the urban facilities which meet any one of target types and meet the scale requirement from all the urban facilities in the set area to serve as external flow targets.
Before the characteristic establishing module establishes the external flowing target characteristic quantity, a flowing target screening module is used for screening flowing targets capable of periodically or all weather attracting non-negligible or even huge personnel individuals from a target set consisting of all targets in a d1 administrative district according to target type requirements and scale requirements to serve as flowing targets meeting the requirements. Wherein the target type comprises at least one of: transportation facilities, medical facilities, residential facilities, commercial and service facilities, educational facilities. That is, the external flow object is required to be one of the above object types, which cover facilities that are all facilities that may attract a large number of individuals to go, such as subway stations and high-speed railway stations in traffic facilities, large comprehensive hospitals in medical facilities, residential districts in residential facilities, shopping centers, business office centers, catessens, parks, movie theaters, factories, etc. in business and service facilities, college cities in educational facilities.
Because the facilities in the same target type have large and small scales, the subway station beside the public transport hub is generally large in scale and strong in flow suction capacity, and the public transport station in a remote suburb is small in scale and extremely weak in flow suction capacity, the scale requirements are met after the target type is met, for example, the daily average traffic throughput can reach 1000 persons, or the facility has large influence in a city, and the like. The scale requirement may depend on the size of the selected area to be set when the process is carried out, and generally the larger the area of the area, the higher the scale requirement.
By screening out facilities with strong individual mobility from city elements, city facilities and city units in a set area as external flow targets, the finally generated layout scheme is closer to the intention of an implementer, and the monitoring effect of the monitoring system is improved.
In one embodiment, the system further comprises a characteristic establishing module, wherein the characteristic establishing module is used for selecting a starting target from the urban facility as an individual flowing starting position and establishing an external flowing target characteristic quantity of the external flowing target in the set area to the individual.
The starting target is the starting position of individual flowing, the external flowing target in the d1 administrative district is used as the target position of individual flowing, and the individual flowing process is completed from the starting target to a certain external flowing target. After the initial target is selected, when the characteristic establishing module establishes the characteristic quantity of the external flowing target, the characteristic quantity is established on the premise that the individual is at the position of the initial target.
The starting target may be selected from urban facilities in the d1 administrative district, for example, from satisfactory external flowing targets screened by the flowing target screening module, and the characteristic quantity of the external flowing target is the characteristic quantity of the remaining external flowing targets in the d1 administrative district to the starting target.
The starting target can be selected from outside the d1 administrative district, but is usually a city facility with strong current suction capacity, and the characteristic quantity of the external flowing target in the current situation is the characteristic quantity of all external flowing targets in the d1 administrative district to the starting target.
The feature establishing module expresses the external flow target feature quantity as:
where i ═ 1, k is the target number of external flows within the set area. If the starting target selected by the feature establishing module is a city facility located outside the d1 administrative district, k is equivalent to the total number m of the floating targets in the d1 administrative district, and if the starting target selected by the feature establishing module is an external floating target in the d1 administrative district, k is equivalent to the number m-1 of the remaining floating targets in the d1 administrative district.
βiThe influence value of the ith external flow target is proportional to the ability of the external flow target to perform the corresponding function. For example, for a subway station, the longer the subway line and the more transit stations, the larger the extension range, which is equivalent to the stronger the ability of the subway station to perform manned transportation to a certain position, and therefore the more the number of degrees of influence thereof will beLarge; otherwise, the smaller. For a shopping center, the larger the parking lot is, the more the number and the coverage types of the resident merchants are, the higher the commodity cost ratio is, and the larger the influence degree value is; otherwise, the smaller.
The distance and the capability of fulfilling corresponding functions are taken as consideration factors to be added into the operation process of the external flow target characteristic quantity, so that the final prediction result is more accurate and closer to reality.
In one embodiment, the system further comprises an individual partitioning module and a model building module. The individual dividing module is used for dividing the types of the individuals according to the individual conditions of the individuals.
Before the tendency degree operation module inputs the external flowing target characteristic quantity into the neural network models of the flowing behaviors of various individuals, the neural network models which can respectively correspond to the individuals with different personal conditions need to be established in advance. Therefore, the individuals at the starting target are divided by the individual dividing module, namely, the types of the individuals are distinguished by utilizing the difference of internal causes. The partition items include at least one of: age, work status, travel mode, income level, family status, health status, consumption demand level. Finally, a plurality of individuals close to the real crowd are obtained. The number of the obtained individual types is the product of the number of the divided grades of each item in the adopted divided items, for example, the age is divided into 4 grades from 12 to 50 years, the income level is divided into 3 grades according to the average monthly income, the consumption demand level is divided into 3 grades according to the average monthly consumption amount, and n is 24 different types of individuals in total.
The types of the individual mobility behavior neural network models are generally more, and for the range of the set area, only partial types of neural network models are generally used, for example, a d1 administrative district is a hot spot area of a first-line city, and mainly takes an emerging technology industry park as a main part, so that the ages are generally younger, the income level and the consumption demand level are generally higher, and therefore, the neural network models of the types of the high ages and the low income and low consumption are placed in the set area in a small number or even not to be placed close to the real situation.
The model establishing module is used for establishing an individual flowing behavior neural network model according to the individual type divided by the individual dividing module as a model type, so that various neural network models can respectively correspond to various individuals with different individual conditions.
By carrying out type division on the individuals according to the real life situations of the individuals, the predicted flowing result is closer to the real situation.
After the model building module builds a plurality of types of individual floating behavior neural network models, a real sample of the floating behavior of a real person is collected, and the model is trained by the sample, so that the output of the floating behavior tendency degree generated by the model is matched with the real behavior of a real floating population individual to the maximum extent, therefore, in one embodiment, the system further comprises a model training module for training the neural network model. The model training module comprises a sample determining unit, a data acquisition unit and a model training unit.
The sample determining unit is used for determining a plurality of individuals with different personal conditions as individual samples. Specifically, 10000 persons, for example, may be collected as samples covering different individual situations, for example, covering different ages, work conditions, travel patterns, income levels, family conditions, health conditions, and consumption demand levels, whose elements correspond to several categories of neural network models, respectively.
The data acquisition unit is used for acquiring the actual flowing behavior of the individual sample determined by the sample determination unit, for example, tracking the strokes of a plurality of individuals, and calculating the external flowing target characteristic quantity corresponding to the actual flowing behavior, for example, taking the residence of the individual as the initial position, calculating the external flowing targets which the individual has passed in a period of time, and calculating the characteristic quantity corresponding to the external flowing targets through the characteristic establishment module.
The model training unit is used for inputting the external flowing target characteristic quantity calculated by the data acquisition unit into the neural network model, and adjusting the neuron parameters of the neural network model based on the matching degree between the flowing behavior tendency degree output by the neural network model and the actual flowing behavior of the corresponding sample individual acquired by the data acquisition unit until the matching degree meets the requirement.
The individual flow behavior neural network model may employ a BP neural network model. The core of the BP neural network model is a BP algorithm, the BP algorithm is carried out according to the direction from input to output when calculating error output, and the weight and the threshold are adjusted according to the direction from output to input. During forward propagation, an input signal acts on an output node through a hidden layer, an output signal is generated through nonlinear transformation, and if actual output does not accord with expected output, the process of backward propagation of errors is carried out. The error back transmission is to back transmit the output error to the input layer by layer through the hidden layer, and to distribute the error to all units of each layer, and to use the error signal obtained from each layer as the basis for adjusting the weight of each unit. The error is reduced along the gradient direction by adjusting the connection strength of the input node and the hidden node, the connection strength of the hidden node and the output node and the threshold value, the network parameters (weight and threshold value) corresponding to the minimum error are determined through repeated learning and training, and the training is stopped immediately.
When the feature building module builds the neural network model and the model training module trains the neural network model, the scene of a specific urban space is not involved, but only a general model is built and trained, for example, for a certain model alone, the tendency of the model to a subway station 3 km away from the station is 80, and the position of the model (namely the initial position of an individual) is not considered, and the specific position of the subway station 3 km away is not considered. Only when the neural network model is put into an actual urban area and the individual mobility distribution is predicted through the system, the influence of stimulation of each large-scale urban facility in the urban area on the neural network model is reflected, and prediction is achieved according to the influence.
In one embodiment, the scheme generation module screens the external flow targets reaching the standard from the external flow targets by using a preset individual number threshold or an individual number ratio threshold, or by using individual number ranking.
The individual number threshold is a criterion that measures whether the flow objective has reached a standard in an absolute numerical manner. For example, if the number of individuals threshold is set to 8000, 5000 individuals flowing between the cell and the P subway station are expected, and 9000 individuals flowing between the cell and the Q shopping mall station are expected, the layout scheme is generated such that the monitoring system is not installed between the cell and the P subway station, but the monitoring system is installed between the cell and the Q shopping mall. The method is suitable for monitoring the flow individual of which the flow quantity only reaches a certain degree, and the flow individual of the flow quantity is required to be monitored, and according to different measuring results, a monitoring system is not required to be installed between any paths, or monitoring systems are required to be installed on all paths.
The individual number ranking is a standard for measuring whether the flowing target reaches the standard or not in a mode of comparing relative sizes. For example, there are 5 flow targets, and the corresponding monitoring system is provided with only 3 sets, so that the cell and the first 3 predicted flow targets with the highest number of flow individuals in the 5 flow targets can be selected as the standard flow targets, no matter what the number of the flow individuals in the first 3 is, and the 3 sets of monitoring systems are respectively installed between the cell and the paths of the standard flow targets. This is suitable for selecting the most important path for installation when the monitoring system needs to be installed anyway.
The individual number ratio threshold is between the two modes, which can be biased to absolute measurement or relative comparison, and the bias determining factor is the size of the individual number ratio threshold. For example, if the number of individuals is set to 30%, the ratio of the total number of individuals flowing between the cell and the P subway station to the total number of sampled individuals is 5000/(5000+9000+8000) to 22.7%, the ratio of the total number of individuals flowing between the cell and the Q shopping center to the total number of sampled individuals is 9000/(5000+9000+8000) to 40.9%, and the ratio of the total number of individuals flowing between the cell and the R office building is 9000/(5000+9000+8000) to 36.4%, the layout plan is created in which the monitoring system is installed only between the cell and the Q shopping center and between the cell and the R office building.
In one embodiment, the layout scheme further includes the installation number of the security monitoring systems, and the installation number is determined at least according to the corresponding number of the mobile individuals or the ratio of the corresponding number of the mobile individuals of each standard external mobile target. Under the condition that the number of the monitoring systems available for installation is limited, when the external flowing target reaching the standard is determined through the individual number threshold and the individual number ranking, the installation number of the monitoring systems is determined according to the corresponding flowing individual number, and when the external flowing target reaching the standard is determined through the individual number ratio threshold, the installation number of the monitoring systems is determined according to the corresponding flowing individual number ratio.
In order to perform more complex and detailed individual mobility prediction in a set area, in one embodiment, the system further includes a prediction data integration module, and the prediction data integration module is configured to integrate the number distribution of the respective flowing individuals generated by the flow prediction module under a plurality of different scenarios to obtain an overall number distribution.
That is to say, after a layout scheme about a starting target a which is located in a set area and is an external flowing target is generated through the scheme generation module, the system can be continuously implemented, and finally, a layout scheme with a larger area range and/or different flowing directions is generated through the scheme generation module, so that individual flowing distribution of an urban area can be observed more comprehensively and in multiple angles, the final overall layout scheme is made to be more macroscopic, the consideration factors are more comprehensive, and the installation of the monitoring system is made to be more reasonable and optimized.
Factors that distinguish whether the scenes are different include: individual flow direction, and/or range of a defined region.
When the scene includes individual flow direction, assuming that the flow target screening module judges that 5 external flow targets exist in the set area, wherein one cell, one business center and one university city exist, and the other two are a subway station and a shopping center, the flow distribution prediction of the individuals can be carried out three times by respectively using the cell, the business center and the university city as starting targets, so as to respectively obtain the number distribution of the flow individuals of the three times, and then the same prediction number of the relevant parties is accumulated. For example, with the cell as the starting target, the neural network models with the corresponding number and types are put in the process according to the morning sampling standard, and the total number of people flowing from the cell to the business center is calculated to be 6000, while with the business center as the starting target, the neural network models with the corresponding number and types are put in the process according to the evening sampling standard, and the total number of people flowing from the business center to the cell is calculated to be 7000. When the scheme generation module judges that the external flow targets reach the standard, the individual flow number is the accumulated value 6000+7000 or the average value (6000+7000)/2, the value calculated according to the accumulated value or the average value or other operation modes depends on the operation mode of other related parties when calculating, and if the related party of the cell-business center adopts the accumulated value, other related parties also adopt the accumulated value.
In this way, errors generated in the prediction process can be balanced, so that the prediction result is closer to the real situation.
When the scene includes the range of the set area, the set area when the system is implemented for the first time for prediction is assumed to be the range of a radius of 5 kilometers with the starting target a as the center of a circle, and includes seven external flow targets a-G, the set area when the system is implemented for the second time for prediction is assumed to be the range of a radius of 5 kilometers with the starting target B as the center of a circle, and includes nine external flow targets a-C, H-M, and the set area when the system is implemented for the third time for prediction is assumed to be the range of a radius of 5 kilometers with the starting target C as the center of a circle, and includes eight external flow targets a-C, N-R. And equivalently, a triangular area formed by A, B, C three flow targets is used as a starting target, and a corresponding monitoring system layout scheme is generated according to flow prediction between the triangular area and the outside in a certain range. It can be understood that, in the process, although the ranges of the set areas selected in each implementation are different, a certain area range intersection is required, and the union of the set areas in each implementation is the whole city space range for the layout of the monitoring system.
In this way, the range of the set area and the emphasis of observation can be customized specifically, and the layout of the monitoring system can be performed more specifically.
The following describes in detail a first embodiment of the urban security monitoring layout method using target mobility distribution prediction according to the present invention with reference to fig. 3, where this embodiment is a method for implementing the urban security monitoring layout system disclosed in the first embodiment of the urban security monitoring layout system.
As shown in fig. 3, the city security monitoring layout system disclosed in this embodiment mainly includes the following steps:
and step 100, inputting the external flowing target characteristic quantity capable of reflecting the influence degree of the external flowing target in the set area on the flow of the individual into various individual flowing behavior neural network models to obtain the flowing behavior tendency of each individual to the corresponding external flowing target. Wherein, each kind of neural network model can correspond multiple individual that individual condition is different respectively.
And 200, predicting the corresponding flowing individual quantity distribution generating flowing behaviors for each external flowing target in the set area based on the flowing behavior tendency degree of each body output by each neural network model to the corresponding external flowing target.
And 300, screening the external mobile targets with the mobile individual number reaching the standard from the external mobile targets, and generating a security monitoring system layout scheme of the corresponding urban area based on the external mobile targets reaching the standard. The layout scheme can at least determine the installation area of the security monitoring system.
In one embodiment, the method further comprises:
step A1: before step 100, urban facilities meeting any one of the target types and the size requirement are screened out from all urban facilities in the set area as external flow targets. Wherein the target type comprises at least one of: transportation facilities, medical facilities, residential facilities, commercial and service facilities, educational facilities.
In one embodiment, the method further comprises:
step A2: after step a1 and before step 100, selecting a starting target from the urban facility as an individual flowing starting position, and establishing an external flowing target characteristic quantity of the external flowing target in the set area to the individual.
In one embodiment, the external flow target characteristic quantity is expressed as:
wherein i ═ 1, k]K is the number of external flow targets within the set area,
for the influence factor of the ith external flow target on the individual, diIs the distance of the path between the ith external flow target and the individual, the influence factor and diIn inverse proportion, βiThe influence value of the ith external flow target is proportional to the ability of the external flow target to perform the corresponding function.
In one embodiment, the method further comprises:
step B1: before step 100, the types of individuals are divided according to the individual conditions of the individuals, and the individual flow behavior neural network models are established according to the divided individual types as model types, so that various neural network models can respectively correspond to the individuals with different individual conditions. Wherein the personal condition comprises at least one of: age, work status, travel mode, income level, family status, health status, consumption demand level.
In one embodiment, the method further comprises:
step B2: after step B1 and before step 100, the neural network model is trained.
Wherein training the neural network model comprises:
several individuals with different personal conditions are determined as individual samples.
The actual flow behavior of the individual sample determined by the sample determination unit is collected, and the external flow target characteristic quantity corresponding to the actual flow behavior is calculated.
And inputting the external flowing target characteristic quantity calculated by the data acquisition unit into the neural network model, and adjusting the neuron parameters of the neural network model based on the matching degree between the flowing behavior tendency degree output by the neural network model and the actual flowing behavior of the corresponding sample individual acquired by the data acquisition unit until the matching degree meets the requirement.
In one embodiment, the individual ambulatory behavior neural network model employs a BP neural network model.
In one embodiment, the qualified external flow target is screened using an individual number threshold or an individual number ranking or an individual number proportion threshold.
In an embodiment, the layout scheme further includes an installation number of the security monitoring methods, and the installation number is determined at least according to the number of the corresponding mobile individuals or the ratio of the number of the corresponding mobile individuals of each standard external mobile target.
In one embodiment, the installation area in the layout scheme is determined in conjunction with map information of the set area.
In one embodiment, the method further comprises:
and C: after step 200, the number distributions of the respective flowing individuals generated under a plurality of different scenarios are integrated to obtain an overall number distribution, and the external flowing target reaching the standard is screened based on the overall number distribution in step 300. Wherein the factors for distinguishing whether the scenes are different include: individual flow direction, and/or range of a defined region.
It should be noted that: the embodiments described are some embodiments of the present invention, not all embodiments, and features in embodiments and embodiments in the present application may be combined with each other without conflict. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The division of modules, units or components herein is merely a logical division, and other divisions may be possible in an actual implementation, for example, a plurality of modules and/or units may be combined or integrated in another system. Modules, units, or components described as separate parts may or may not be physically separate. The components displayed as cells may or may not be physical cells, and may be located in a specific place or distributed in grid cells. Therefore, some or all of the units can be selected according to actual needs to implement the scheme of the embodiment.
The above description is only for the specific embodiments of the present invention, but the scope of the present invention is not limited thereto, and any changes or substitutions that can be easily conceived by those skilled in the art within the technical scope of the present invention are included in the scope of the present invention. Therefore, the protection scope of the present invention shall be subject to the protection scope of the appended claims.
Claims (8)
1. The utility model provides an adopt city security protection monitoring layout system of target mobility distribution prediction which characterized in that includes:
the tendency calculation module is used for inputting the characteristic quantity of the external flowing target capable of reflecting the influence degree of the external flowing target in the set area on the flow of the individual into various individual flowing behavior neural network models to obtain the flowing behavior tendency of each individual to the corresponding external flowing target;
the flow predicting module is used for predicting the corresponding flow individual quantity distribution of flow behavior generated on each external flow target in the set area based on the flow behavior tendency degree;
the scheme generation module is used for screening out external mobile targets with the mobile individual number reaching the standard and generating a security monitoring system layout scheme of the corresponding urban area based on the external mobile targets reaching the standard and the initial target;
the various neural network models can respectively correspond to various individuals with different personal conditions, and the layout scheme can at least determine the installation area of the security monitoring system;
the system further comprises: the characteristic establishing module is used for selecting an initial target from urban facilities as an individual flowing initial position and establishing the characteristic quantity of the external flowing target in the set area to the individual;
the scheme generation module screens the up-to-standard external flowing target by utilizing an individual quantity threshold value or an individual quantity ranking or an individual quantity ratio threshold value;
the layout scheme further comprises the installation quantity of the security monitoring systems, and the installation quantity is determined at least according to the corresponding flowing individual quantity or the ratio of the corresponding flowing individual quantity of each standard external flowing target;
the plan generation module determines the installation area in the layout plan at least in conjunction with map information of the set area.
2. The system of claim 1, further comprising:
the flow target screening module is used for screening out urban facilities which meet any one of target types and meet the scale requirement from all the urban facilities in the set area as the external flow targets; wherein the content of the first and second substances,
the target type includes at least one of: transportation facilities, medical facilities, residential facilities, commercial and service facilities, educational facilities.
3. The system of claim 1, wherein the external flow target characteristic quantity is expressed as:
wherein i = [1, k =]K is the target number of external flows within the set area,
for the impact factor of the ith said external flow target on the individual,
is the distance of the path between the ith external flow target and the individual, the influence factor and
in the inverse proportion,
an influence magnitude for the ith external flow target itself, the influence magnitude being proportional to the ability of the external flow target to perform the corresponding function.
4. The system of claim 1, further comprising:
the prediction data integration module is used for integrating the corresponding flow individual quantity distribution generated by the flow prediction module under a plurality of different scenes to obtain integral quantity distribution; wherein the content of the first and second substances,
factors that distinguish whether the scenarios are different include: individual flow direction, and/or extent of the defined region.
5. A city security monitoring layout method adopting target mobility distribution prediction is characterized by comprising the following steps:
inputting the characteristic quantity of the external flowing target capable of reflecting the influence degree of the external flowing target in the set area on the flow of the individual into various individual flowing behavior neural network models to obtain the flowing behavior tendency of each individual to the corresponding external flowing target;
predicting the corresponding flow individual quantity distribution of flow behavior generated on each external flow target in the set area based on the flow behavior tendency degree;
screening out external mobile targets with the mobile individual number reaching the standard, and generating a security monitoring system layout scheme of a corresponding urban area based on the external mobile targets reaching the standard and an initial target;
the various neural network models can respectively correspond to various individuals with different personal conditions, and the layout scheme can at least determine the installation area of the security monitoring system; and the number of the first and second electrodes,
the method further comprises the following steps: selecting an initial target from urban facilities as an individual flowing initial position, and establishing the external flowing target characteristic quantity of the external flowing target in the set area to the individual; in addition, the first and second substrates are,
screening the external flowing target reaching the standard by utilizing an individual quantity threshold value or an individual quantity ranking or an individual quantity ratio threshold value;
the layout scheme further comprises the installation quantity of the security monitoring method, and the installation quantity is determined at least according to the corresponding flowing individual quantity or the ratio of the corresponding flowing individual quantity of each standard external flowing target;
and determining the installation area in the layout scheme by combining the map information of the set area.
6. The method of claim 5, further comprising:
screening out urban facilities which meet any one of target types and meet the scale requirement from all the urban facilities in the set area as the external flowing target; wherein the content of the first and second substances,
the target type includes at least one of: transportation facilities, medical facilities, residential facilities, commercial and service facilities, educational facilities.
7. The method according to claim 5, wherein the external flow target characteristic quantity is expressed as:
wherein i = [1, k ], k is the target number of external flows within the set area,
8. The method of claim 5, further comprising:
integrating the number distribution of the corresponding mobile individuals generated under a plurality of different scenes to obtain integral number distribution, and screening the external mobile target reaching the standard based on the integral number distribution; wherein the content of the first and second substances,
factors that distinguish whether the scenarios are different include: individual flow direction, and/or extent of the defined region.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910453560.5A CN110309953B (en) | 2019-05-28 | 2019-05-28 | Urban security monitoring layout system and method adopting target mobility distribution prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201910453560.5A CN110309953B (en) | 2019-05-28 | 2019-05-28 | Urban security monitoring layout system and method adopting target mobility distribution prediction |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN110309953A CN110309953A (en) | 2019-10-08 |
| CN110309953B true CN110309953B (en) | 2020-06-26 |
Family
ID=68075095
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201910453560.5A Active CN110309953B (en) | 2019-05-28 | 2019-05-28 | Urban security monitoring layout system and method adopting target mobility distribution prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN110309953B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112218046B (en) * | 2020-09-27 | 2023-10-24 | 杭州海康威视系统技术有限公司 | Object monitoring method and device |
| CN115174608A (en) * | 2022-05-25 | 2022-10-11 | 安徽超清科技股份有限公司 | Smart city security monitoring system based on Internet of things |
Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106097222A (en) * | 2016-06-12 | 2016-11-09 | 河南理工大学 | The communal facility spatial accessibility taking user characteristics into account analyzes method and device |
| CN106503829A (en) * | 2016-09-26 | 2017-03-15 | 同济大学 | A kind of crowding Forecasting Methodology of the Urban Public Open Space based on multi-source data |
| CN109583942A (en) * | 2018-11-07 | 2019-04-05 | 浙江工业大学 | A kind of multitask convolutional neural networks customer behavior analysis method based on dense network |
| EP3486769A1 (en) * | 2017-11-20 | 2019-05-22 | Guangdong OPPO Mobile Telecommunications Corp., Ltd. | Method for preloading application, terminal device, and medium |
Family Cites Families (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN102332122A (en) * | 2011-10-18 | 2012-01-25 | 东南大学 | Layout optimization method for urban public bicycle rental stations |
| KR101830522B1 (en) * | 2016-08-22 | 2018-02-21 | 가톨릭대학교 산학협력단 | Method for predicting crime occurrence of prediction target region using big data |
| CN108021980B (en) * | 2017-12-15 | 2020-04-17 | 中国科学院地理科学与资源研究所 | Fine-scale urban population quantity prediction method based on mobile phone data |
| CN109376935B (en) * | 2018-10-31 | 2021-07-02 | 东南大学 | Bus passenger flow time interval combined prediction method based on neural network |
| CN109740292A (en) * | 2019-01-30 | 2019-05-10 | 中国测绘科学研究院 | A kind of urban population spatial distribution evaluation method and device based on multiple agent |
-
2019
- 2019-05-28 CN CN201910453560.5A patent/CN110309953B/en active Active
Patent Citations (4)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106097222A (en) * | 2016-06-12 | 2016-11-09 | 河南理工大学 | The communal facility spatial accessibility taking user characteristics into account analyzes method and device |
| CN106503829A (en) * | 2016-09-26 | 2017-03-15 | 同济大学 | A kind of crowding Forecasting Methodology of the Urban Public Open Space based on multi-source data |
| EP3486769A1 (en) * | 2017-11-20 | 2019-05-22 | Guangdong OPPO Mobile Telecommunications Corp., Ltd. | Method for preloading application, terminal device, and medium |
| CN109583942A (en) * | 2018-11-07 | 2019-04-05 | 浙江工业大学 | A kind of multitask convolutional neural networks customer behavior analysis method based on dense network |
Also Published As
| Publication number | Publication date |
|---|---|
| CN110309953A (en) | 2019-10-08 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Vlahogianni et al. | A real-time parking prediction system for smart cities | |
| Zhong et al. | Inferring building functions from a probabilistic model using public transportation data | |
| CA2561939C (en) | Method and system for forecasting events and results based on geospatial modeling | |
| US7801842B2 (en) | Method and system for spatial behavior modification based on geospatial modeling | |
| Huang et al. | A mobility network approach to identify and anticipate large crowd gatherings | |
| US7571146B2 (en) | Event, threat and result change detection system and method | |
| Bao et al. | Spatial analysis of bikeshare ridership with smart card and POI data using geographically weighted regression method | |
| CN114897444B (en) | Method and system for identifying service facility requirements in urban subarea | |
| Le Bescond et al. | Open-source modeling chain for the dynamic assessment of road traffic noise exposure | |
| Pulugurtha et al. | Modeling annual average daily traffic with integrated spatial data from multiple network buffer bandwidths | |
| Chen et al. | Multi-step prediction of experienced travel times using agent-based modeling | |
| Lu et al. | Activity-based air pollution exposure assessment: differences between homemakers and cycling commuters | |
| CN111783295B (en) | Dynamic identification and prediction evaluation method and system for urban community specific human behavior chain | |
| CN110309953B (en) | Urban security monitoring layout system and method adopting target mobility distribution prediction | |
| Abenoza et al. | What is the role of weather, built-environment and accessibility geographical characteristics in influencing travelers’ experience? | |
| Beura et al. | Defining Bicycle Levels of Service Criteria Using Levenberg–Marquardt and Self-organizing Map Algorithms | |
| Wang | Improved annual average daily traffic (AADT) estimation for local roads using parcel-level travel demand modeling | |
| Huang et al. | Understanding spatio-temporal mobility patterns for seniors, child/student and adult using smart card data | |
| Xia | Modelling the spatial-temporal movement of tourists | |
| Liu et al. | Short-term online taxi-hailing demand prediction based on the multimode traffic data in metro station areas | |
| Lian et al. | Advances in estimating pedestrian measures through artificial intelligence: From data sources, computer vision, video analytics to the prediction of crash frequency | |
| Kaziyeva et al. | Large-scale agent-based simulation model of pedestrian traffic flows | |
| Jonker | Modelling the trip length distribution of shopping trips from GPS data | |
| Borhani et al. | Soft computing modelling of urban evolution: Tehran metropolis | |
| Jena et al. | Assessment of service quality at unsignalized intersections using traffic simulation and computational intelligence |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |