CN114255588A - Regional motor vehicle illegal parking prediction method, storage medium and device - Google Patents
Regional motor vehicle illegal parking prediction method, storage medium and device Download PDFInfo
- Publication number
- CN114255588A CN114255588A CN202111473572.8A CN202111473572A CN114255588A CN 114255588 A CN114255588 A CN 114255588A CN 202111473572 A CN202111473572 A CN 202111473572A CN 114255588 A CN114255588 A CN 114255588A
- Authority
- CN
- China
- Prior art keywords
- illegal parking
- data
- regional
- parking
- illegal
- 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
Images
Classifications
-
- 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
-
- 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
- G08G1/0108—Measuring and analyzing of parameters relative to traffic conditions based on the source of data
-
- 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
Landscapes
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a regional motor vehicle illegal parking prediction method, a storage medium and equipment, wherein the method comprises the following steps of dividing grids on a managed map according to the step length of one kilometer, wherein one grid is a region; carrying out data preprocessing and association on the relevant data of the region to obtain region summarized data; establishing an illegal parking number model by using a full-connection neural network algorithm according to the acquired regional summary data; forecasting the illegal parking number of the current time period area by using an illegal parking number model through current data; and drawing a thermodynamic diagram by using the predicted number of the illegal parking vehicles in the area so as to see the illegal parking number in a certain area. The invention provides a method for predicting the number of illegal parking of regional motor vehicles, which can predict the number of illegal parking through calculation and process traffic safety hidden dangers more quickly and efficiently. The method is favorable for traffic managers to effectively supervise illegal parking of regional motor vehicles.
Description
Technical Field
The invention relates to the technical field of motor vehicle illegal parking prediction, in particular to a regional motor vehicle illegal parking prediction method, a storage medium and equipment.
Background
In order to effectively prevent traffic accidents and strengthen the management of vehicles, the main functions of enterprises need to be fully exerted, and the source management work of the vehicles needs to be strengthened. Illegal parking can seriously affect road traffic safety and city image, easily cause traffic incidents such as road traffic jam and the like, and greatly reduce the traffic efficiency of vehicles.
Disclosure of Invention
The invention provides a regional motor vehicle illegal parking prediction method, a storage medium and equipment, which utilize the application of a full-connection neural network in a prediction model, provide the regional motor vehicle illegal parking prediction method and are beneficial to the effective supervision of a traffic department on regional illegal parking.
In order to achieve the purpose, the invention adopts the following technical scheme:
a regional motor vehicle illegal parking prediction method is implemented by computer equipment,
dividing grids on a managed map according to the step length of one kilometer, wherein one grid is an area;
carrying out data preprocessing and association on the relevant data of the region to obtain region summarized data;
establishing an illegal parking number model by using a full-connection neural network algorithm according to the acquired regional summary data;
forecasting the illegal parking number of the current time period area by using an illegal parking number model through current data;
and drawing a thermodynamic diagram by using the predicted number of the illegal parking vehicles in the area so as to see the illegal parking number in a certain area.
Further, the area related data comprises area parking spaces, area residential cells, historical area traffic flow per time period and business circles.
Further, the data preprocessing and association are performed on the relevant data of the region to obtain region summarized data, and the method specifically includes:
2.1, counting the environment in the region and the surrounding environment, counting the basic environment condition in the grid region, and if N cells exist in the region, the adjacent grid regions have M residential cells; then the residential cell of the grid area is N + M/2;
if the grid area has a business circle, marking the business circle as 1, and if the adjacent grid area has the business circle, marking the business circle as 0.5, and if no business circle is 0; at the moment, the grid area table comprises grid serial numbers, upper left coordinates, upper right coordinates, lower left coordinates, lower right coordinates, cell number and business circles;
2.2, acquiring parking lot data according to the parking lot table;
acquiring 'longitude of center point', 'latitude of center point', 'parking space' of the parking lot according to the parking lot table, and associating with the 2.1-step grid area table to obtain grid serial number, upper left coordinate, upper right coordinate, lower left coordinate, lower right coordinate, cell number and parking space;
2.3, counting point traffic data to obtain regional real-time and historical traffic flow;
the historical traffic flow statistics is carried out on the traffic flow of each hour time period of the last year; the current field of the data is a point location number, point location precision, point location dimension, time period and traffic flow;
2.4, acquiring historical violation data;
and acquiring illegal data of illegal parking from the historical illegal record table, acquiring illegal time and illegal places, associating grid area data according to longitude and latitude of the illegal places, and summarizing according to grid serial numbers and time periods to obtain historical illegal parking data which are grid serial numbers, time periods and illegal parking quantity.
Further, the establishing of the illegal parking number model by using a full-connection neural network algorithm according to the obtained regional summary data specifically comprises processing data according to the obtained characteristics of the regional environment, the parking lot, the historical time period traffic flow and the historical illegal parking data, and establishing the illegal parking number model by using the full-connection neural network;
the illegal parking digital model adopts a neural network model as follows:
comprises an input layer, one or more hidden layers and an output layer; the input layer takes five indexes of regional parking spaces, historical regional vehicle flow, time periods, residential districts and commercial districts as input, the hidden layer is used for calculation, and the output layer takes the regional historical number of violations as output;
the number of nodes of the input layer is 5, the number of nodes of the hidden layer is 7, and the number of nodes of the output layer is 1.
Further, the illegal parking digital model training process firstly generates W randomly1，W2Two parameters, input data calculate output data y through forward propagation, and then calculate loss value loss through loss function by using the output data and target value y';
optimizing the loss value by using back propagation; backward propagation uses the gradient descent principle to update the parameter W step by step forward1，W2Until the loss value is minimal.
Further, the forward propagation calculation process is as follows:
an input layer: x ═ X1, X2, X3, X4, X5
Inputting layer parameters:
input layer bias term: b1＝[b11，b12，b13，b14，b15，b16，b17]
Hidden layer parameters:
hidden layer bias term: b2＝b2
The activation function is a RELU function: f (x) max (x, 0)
The hidden layer output is:
X×W1+b1＝A＝[f(a1)，f(a2)，f(a3)，f(a4)，f(a5)，f(a6)，f(a7)]
the output layer is: f (A × W)2+b2)＝y。
Further, the loss function of the illegal parking number model is as follows:
for a training sample, assuming that the output of the model is y and the target value of the training sample is y ', the loss value of the single sample is | y' -y |;
for multiple samples, using the mean square error, then the loss value at this time is:
further, the back propagation process is as follows:
taking the mean square error as a loss function, there are:
in the above function, X is given, and a random value W is initialized1、W2Calculating A and y, and updating parameter W by gradient descent method2A value of (d);
the above loss function continues to be expanded to:
further, a loss function related to the independent variable W is obtained, and then the parameter W is updated by using gradient descent.
In yet another aspect, the present invention also discloses a computer readable storage medium storing a computer program, which when executed by a processor causes the processor to perform the steps of the method as described above.
In yet another aspect, the present invention also discloses a computer device comprising a memory and a processor, the memory storing a computer program which, when executed by the processor, causes the processor to perform the steps of the above method.
According to the technical scheme, the regional motor vehicle illegal parking number prediction method analyzes the regional environment, the real-time traffic flow, the historical illegal parking situation, the parking lot parking spaces and other data through the application of the full-connection neural network in the regression prediction model, and establishes the regional motor vehicle illegal parking number prediction model.
The invention provides a method for predicting the number of illegal parking of regional motor vehicles, which can predict the number of illegal parking through calculation and process traffic safety hidden dangers more quickly and efficiently. The method is favorable for traffic managers to effectively supervise illegal parking of regional motor vehicles.
Drawings
FIG. 1 is a block flow diagram of the present invention;
FIG. 2 is a model architecture diagram of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention.
As shown in fig. 1, the method for predicting illegal parking of regional motor vehicles according to the embodiment performs the following steps by using a computer device,
dividing grids on a managed map according to the step length of one kilometer, wherein one grid is an area;
carrying out data preprocessing and association on the relevant data of the region to obtain region summarized data;
establishing an illegal parking number model by using a full-connection neural network algorithm according to the acquired regional summary data;
forecasting the illegal parking number of the current time period area by using an illegal parking number model through current data;
and drawing a thermodynamic diagram by using the predicted number of the illegal parking vehicles in the area so as to see the illegal parking number in a certain area.
The following are exemplified:
1. and carrying out grid division on the managed map according to the step length of one kilometer, wherein one grid is an area.
For example, four points of longitude and latitude (104.62, 30.16), (104.63, 30.16), (104.62, 30.15), (104.63, 30.15) form a grid. The divided form data are grid serial number, upper left coordinate, upper right coordinate, lower left coordinate and lower right coordinate.
2. Carrying out data preprocessing and association on the relevant data of the region to obtain region summarized data; the regional related data comprise characteristics such as regional parking spaces, regional residential districts, regional traffic flow of each historical time period, business circles and the like; the following is a detailed description:
2.1, statistics of the environment in the area and the surrounding area, such as several residential districts and whether business circles exist. And (4) counting the basic situation of the environment in the grid area, and if N cells exist in the area, the adjacent grid area has M residential cells. Then the residential cell of the grid area is N + M/2. If the grid area has a business circle, the business circle is marked as 1, if the adjacent grid area has a business circle, the business circle is marked as 0.5, and the non-business circle is 0. At this time, the grid area table is a grid serial number, an upper left coordinate, an upper right coordinate, a lower left coordinate, a lower right coordinate, the number of cells and the existence of a business circle.
And 2.2, acquiring parking lot data according to the parking lot table.
And acquiring the longitude of a central point, the latitude of the central point and the parking space of the parking lot according to the parking lot table, and associating the longitude of the central point, the latitude of the central point and the parking space with the 2.1-step grid area table to obtain a grid serial number, an upper left coordinate, an upper right coordinate, a lower left coordinate, a lower right coordinate, the number of cells and parking spaces.
And 2.3, counting point traffic data to obtain regional real-time and historical traffic flow.
The historical traffic flow counts the traffic flow for each hour period of the last year. And the current field of the data is a point location number, point location precision, point location dimension, time period and traffic flow.
2.4 obtaining historical violation data
And acquiring illegal data of illegal parking from the historical illegal record table, acquiring illegal time and illegal places, associating grid area data according to longitude and latitude of the illegal places, and summarizing according to grid serial numbers and time periods to obtain historical illegal parking data which are grid serial numbers, time periods and illegal parking quantity.
3. Establishing an illegal parking number model by using a full-connection neural network algorithm according to the acquired regional summary data;
and processing the data according to the acquired characteristics of the regional environment, the parking lot, the historical time period traffic flow and the historical illegal parking data, and establishing an illegal parking number model by using a full-connection neural network.
The algorithm model is calculated as follows:
3.1 fully-connected neural network model
The neural network consists of three parts: input layer, hidden layer(s), output layer. The model input layer of the invention takes five indexes of regional parking space, historical regional traffic flow, time period, residential district and business district as input, the hidden layer is used for calculation, and the output layer takes regional historical illegal parking number as output. The number of nodes of the input layer of the neural network is 5, the number of nodes of the hidden layer is 7, and the number of nodes of the output layer is 1.
The model training process first randomly generates W1，W2Two parameters, input data, calculate output data y by forward propagation, and then calculate loss value loss by loss function using the output data and target value y'. And when the loss value is minimum, the model is considered to be optimal. The loss value is then optimized by back propagation. Backward propagation uses the gradient descent principle to update the parameter W step by step forward1，W2Until the loss value is minimal. The principle related to the related method is as follows:
3.2, Forward propagation
An input layer: x ═ X1, X2, X3, X4, X5
Inputting layer parameters:
input layer bias term: b1＝[b11，b12，b13，b14，b15，b16，b17]
Hidden layer parameters:
hidden layer bias term: b2＝b2
The activation function is a RELU function: f (x) max (x, 0)
The hidden layer output is: x is W1+b1＝A＝[f(a1)，f(a2)，f(a3)，f(a4)，f(a5)，f(a6)，f(a7)]
The output layer is: f (A × W)2+b2)＝y
3.3 loss function
In machine learning, a loss function is used to measure the loss (gap) between a model output value and a target value. For example: in the neural network, for a training sample, assuming that the output of the model is y and the target value of the training sample is y ', the loss value of the single sample is | y' -y |. For multiple samples, for the convenience of derivation and calculation, the mean square error is generally used, and then the loss value at this time is:
since X is known, where y' can be expressed as the output of the neural network parameter W, loss can be expressed as an argument as a function of W. The goals of the machine learning training at this time are: find the parameter W that minimizes the loss value. This process is the training process of machine learning.
3.4 gradient descent
In neural networks, the most common optimization method is the back propagation algorithm. The training process of the neural network is as follows: the value of the parameter W is solved such that the loss value of the loss function is as small as possible (it is generally difficult to solve for the minimum). The neural network training optimizes the value of the parameter W.
The back propagation algorithm is based on a gradient descent method, and the process of the gradient descent method is simple, namely: and continuously and iteratively updating the parameters to enable the loss value to continuously move towards the direction of the minimum value. The gradient here represents the gradient (slope, first derivative) of the function, given a parameter value, such that the parameter value is constantly moving in the opposite direction of the gradient (slope) so that the parameter value is constantly approaching a minimum.
In training, there will be multiple parameter values, such as W1，W2. When using gradient descent optimization, firstInitial values are randomly generated for them and then each parameter value is updated in a round of learning. Such as updating W1When it comes, all parameters except it are fixed, and then W is updated1. Then W is updated2While still using W1Updating the value before updating, fixing other parameters, and then updating W2. After all the W are updated, the learning (updating) for one round is finished, and then the next round of learning (updating) is entered.
3.5, counter-propagating
Back propagation is the most commonly used optimization algorithm in neural networks, the principle of which is based on gradient descent for optimizing parameters in neural networks. In short, starting from the output layer, the parameters of each layer are updated gradually and forwards in a gradient descending manner. This direction is opposite to the forward propagation direction and is therefore called counter-propagation. Here we describe the process of back propagation:
taking the mean square error as a loss function, there are:
in the above function, X is given, and a random value W is initialized1、W2Obtaining A and y, and updating parameter W by gradient descent method2The value of (c). The above loss function continues to be expanded to:
a further loss function on the argument W can be obtained and then the parameter W is updated using the gradient descent.
4. And predicting the number of illegal parking in the current time period area through the model.
4.1, obtaining the current data
And counting the traffic flow of nearly 1 hour according to the point positions from the traffic passing meter and calculating the time period of the traffic flow. And the current field of the data is point location number, point location precision, point location dimensionality and traffic flow. And (4) acquiring grid area data such as regional parking spaces, residential districts, business circles and the like according to the step 2.
4.2, prediction
And 3, substituting the current data into the model through the established illegal parking regression model in the step 3 to obtain the prediction result of the number of illegal parking vehicles in the current area.
5. Thermodynamic diagrams are drawn by predicting illegal parking vehicles.
The number of parked cars in each area can be obtained after prediction. By drawing a thermodynamic diagram by using the data, a traffic police can obviously see what areas have more current parking lots through the thermodynamic diagram, and can distribute police force to manage in time.
In summary, the regional motor vehicle illegal parking number prediction method provided by the invention analyzes the regional environment, real-time traffic flow, historical illegal parking situation, parking lot parking space and other data through the application of the fully-connected neural network in the regression prediction model, and establishes the regional motor vehicle illegal parking number prediction model. The invention can predict the number of illegal parking through calculation and process the traffic safety hidden danger more quickly and efficiently. The method is favorable for traffic managers to effectively supervise illegal parking of regional motor vehicles.
It is understood that the system provided by the embodiment of the present invention corresponds to the method provided by the embodiment of the present invention, and the explanation, the example and the beneficial effects of the related contents can refer to the corresponding parts in the method.
The embodiment of the application also provides an electronic device, which comprises a processor, a communication interface, a memory and a communication bus, wherein the processor, the communication interface and the memory complete mutual communication through the communication bus,
a memory for storing a computer program;
the processor is used for realizing the method for predicting the illegal parking of the motor vehicle in the area when executing the program stored in the memory;
the communication bus mentioned in the electronic device may be a Peripheral Component Interconnect (PCI) bus or an Extended Industry Standard Architecture (EISA) bus. The communication bus may be divided into an address bus, a data bus, a control bus, etc.
The communication interface is used for communication between the electronic equipment and other equipment.
The Memory may include a Random Access Memory (RAM) or a Non-Volatile Memory (NVM), such as at least one disk Memory. Optionally, the memory may also be at least one memory device located remotely from the processor.
The Processor may be a general-purpose Processor, including a Central Processing Unit (CPU), a Network Processor (NP), and the like; the Integrated Circuit may also be a Digital Signal Processor (DSP), an Application Specific Integrated Circuit (ASIC), a Field Programmable Gate Array (FPGA), or other Programmable logic devices, discrete Gate or transistor logic devices, or discrete hardware components.
In yet another embodiment provided by the present application, a computer-readable storage medium is further provided, in which a computer program is stored, and the computer program, when executed by a processor, implements the steps of any of the above-mentioned methods for predicting illegal parking of a motor vehicle in a region.
In yet another embodiment provided by the present application, there is also provided a computer program product containing instructions that, when executed on a computer, cause the computer to perform the method for predicting illegal parking of a regional vehicle in any of the above embodiments.
In the above embodiments, the implementation may be wholly or partially realized by software, hardware, firmware, or any combination thereof. When implemented in software, may be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When loaded and executed on a computer, cause the processes or functions described in accordance with the embodiments of the application to occur, in whole or in part. The computer may be a general purpose computer, a special purpose computer, a network of computers, or other programmable device. The computer instructions may be stored in a computer readable storage medium or transmitted from one computer readable storage medium to another, for example, from one website site, computer, server, or data center to another website site, computer, server, or data center via wired (e.g., coaxial cable, fiber optic, Digital Subscriber Line (DSL)) or wireless (e.g., infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device, such as a server, a data center, etc., that incorporates one or more of the available media. The usable medium may be a magnetic medium (e.g., floppy Disk, hard Disk, magnetic tape), an optical medium (e.g., DVD), or a semiconductor medium (e.g., Solid State Disk (SSD)), among others.
It is noted that, herein, relational terms such as first and second, and the like may be used solely to distinguish one entity or action from another entity or action without necessarily requiring or implying any actual such relationship or order between such entities or actions. Also, the terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus. Without further limitation, an element defined by the phrase "comprising an … …" does not exclude the presence of other identical elements in a process, method, article, or apparatus that comprises the element.
All the embodiments in the present specification are described in a related manner, and the same and similar parts among the embodiments may be referred to each other, and each embodiment focuses on the differences from the other embodiments. In particular, for the system embodiment, since it is substantially similar to the method embodiment, the description is simple, and for the relevant points, reference may be made to the partial description of the method embodiment.
The above examples are only intended to illustrate the technical solution of the present invention, but not to limit it; although the present invention has been described in detail with reference to the foregoing embodiments, it will be understood by those of ordinary skill in the art that: the technical solutions described in the foregoing embodiments may still be modified, or some technical features may be equivalently replaced; and such modifications or substitutions do not depart from the spirit and scope of the corresponding technical solutions of the embodiments of the present invention.
Claims (10)
1. A regional motor vehicle illegal parking prediction method is characterized by comprising the following steps,
dividing grids on a managed map according to the step length of one kilometer, wherein one grid is an area;
carrying out data preprocessing and association on the relevant data of the region to obtain region summarized data;
establishing an illegal parking number model by using a full-connection neural network algorithm according to the acquired regional summary data;
forecasting the illegal parking number of the current time period area by using an illegal parking number model through current data;
and drawing a thermodynamic diagram by using the predicted number of the illegal parking vehicles in the area so as to see the illegal parking number in a certain area.
2. The regional motor vehicle illegal parking prediction method according to claim 1, characterized in that: the area related data comprises area parking spaces, area residential districts, historical area traffic flow of each time period and business circles.
3. The regional motor vehicle illegal parking prediction method according to claim 2, characterized in that: the method comprises the following steps of performing data preprocessing and association on the related data of the region to obtain region summarized data, wherein the data preprocessing and association specifically comprise the following steps:
2.1, counting the environment in the region and the surrounding environment, counting the basic environment condition in the grid region, and if N cells exist in the region, the adjacent grid regions have M residential cells; then the residential cell of the grid area is N + M/2;
if the grid area has a business circle, marking the business circle as 1, and if the adjacent grid area has the business circle, marking the business circle as 0.5, and if no business circle is 0; at the moment, the grid area table comprises grid serial numbers, upper left coordinates, upper right coordinates, lower left coordinates, lower right coordinates, cell number and business circles;
2.2, acquiring parking lot data according to the parking lot table;
acquiring 'longitude of center point', 'latitude of center point', 'parking space' of the parking lot according to the parking lot table, and associating with the 2.1-step grid area table to obtain grid serial number, upper left coordinate, upper right coordinate, lower left coordinate, lower right coordinate, cell number and parking space;
2.3, counting point traffic data to obtain regional real-time and historical traffic flow;
the historical traffic flow statistics is carried out on the traffic flow of each hour time period of the last year; the current field of the data is a point location number, point location precision, point location dimension, time period and traffic flow;
2.4, acquiring historical violation data;
and acquiring illegal data of illegal parking from the historical illegal record table, acquiring illegal time and illegal places, associating grid area data according to longitude and latitude of the illegal places, and summarizing according to grid serial numbers and time periods to obtain historical illegal parking data which are grid serial numbers, time periods and illegal parking quantity.
4. The regional motor vehicle illegal parking prediction method according to claim 3, characterized in that: establishing an illegal parking number model by using a full-connection neural network algorithm according to the acquired regional summary data, specifically processing data according to the acquired characteristics of regional environment, parking lot, historical time period traffic flow and historical illegal parking data, and establishing the illegal parking number model by using a full-connection neural network;
the illegal parking digital model adopts a neural network model as follows:
comprises an input layer, one or more hidden layers and an output layer; the input layer takes five indexes of regional parking spaces, historical regional vehicle flow, time periods, residential districts and commercial districts as input, the hidden layer is used for calculation, and the output layer takes the regional historical number of violations as output;
the number of nodes of the input layer is 5, the number of nodes of the hidden layer is 7, and the number of nodes of the output layer is 1.
5. The regional motor vehicle illegal parking prediction method according to claim 4, characterized in that:
firstly, randomly generating W in the illegal parking digital model training process1，W2Two parameters, input data calculate output data y through forward propagation, and then calculate loss value loss through loss function by using the output data and target value y';
optimizing the loss value by using back propagation; backward propagation uses the gradient descent principle to update the parameter W step by step forward1，W2Until the loss value is minimal.
6. The regional vehicle illegal parking prediction method according to claim 5, characterized in that:
the forward propagation calculation process is as follows:
an input layer: x ═ X1, X2, X3, X4, X5
Inputting layer parameters:
input layer bias term: b1＝[b11，b12，b13，b14，b15，b16，b17]
Hidden layer parameters:
hidden layer bias term: b2＝b2
The activation function is a RELU function: f (x) max (x, 0)
The hidden layer output is:
X×W1+b1＝A＝[f(a1)，f(a2)，f(a3)，f(a4)，f(a5)，f(a6)，f(a7)]
the output layer is: f (A × W)2+b2)＝y。
7. The regional vehicle illegal parking prediction method according to claim 6, characterized in that:
the loss function of the illegal parking number model is as follows:
for a training sample, assuming that the output of the model is y and the target value of the training sample is y ', the loss value of the single sample is | y' -y |;
for multiple samples, using the mean square error, then the loss value at this time is:
8. the regional vehicle illegal parking prediction method according to claim 6, characterized in that:
the back propagation process is as follows:
taking the mean square error as a loss function, there are:
in the above function, X is given, and a random value W is initialized1、W2Calculating A and y, and updating parameter W by gradient descent method2A value of (d);
the above loss function continues to be expanded to:
further, a loss function related to the independent variable W is obtained, and then the parameter W is updated by using gradient descent.
9. A computer-readable storage medium, storing a computer program which, when executed by a processor, causes the processor to carry out the steps of the method according to any one of claims 1 to 8.
10. A computer device comprising a memory and a processor, the memory storing a computer program that, when executed by the processor, causes the processor to perform the steps of the method according to any one of claims 1 to 8.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111473572.8A CN114255588A (en) | 2021-11-29 | 2021-11-29 | Regional motor vehicle illegal parking prediction method, storage medium and device |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202111473572.8A CN114255588A (en) | 2021-11-29 | 2021-11-29 | Regional motor vehicle illegal parking prediction method, storage medium and device |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN114255588A true CN114255588A (en) | 2022-03-29 |
Family
ID=80793938
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202111473572.8A Pending CN114255588A (en) | 2021-11-29 | 2021-11-29 | Regional motor vehicle illegal parking prediction method, storage medium and device |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN114255588A (en) |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101808203B1 (en) * | 2017-05-10 | 2017-12-12 | (주)서광시스템 | System for cracking vehicle using adaptive predict analysis and method therefor |
| CN109543909A (en) * | 2018-11-27 | 2019-03-29 | 平安科技（深圳）有限公司 | Prediction technique, device and the computer equipment of vehicle caseload |
| CN110619750A (en) * | 2019-08-15 | 2019-12-27 | 重庆特斯联智慧科技股份有限公司 | Intelligent aerial photography identification method and system for illegal parking vehicle |
| CN112016735A (en) * | 2020-07-17 | 2020-12-01 | 厦门大学 | Patrol route planning method and system based on traffic violation hotspot prediction and readable storage medium |
| CN113269042A (en) * | 2021-04-25 | 2021-08-17 | 安徽银徽科技有限公司 | Intelligent traffic management method and system based on running vehicle violation identification |
-
2021
- 2021-11-29 CN CN202111473572.8A patent/CN114255588A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| KR101808203B1 (en) * | 2017-05-10 | 2017-12-12 | (주)서광시스템 | System for cracking vehicle using adaptive predict analysis and method therefor |
| CN109543909A (en) * | 2018-11-27 | 2019-03-29 | 平安科技（深圳）有限公司 | Prediction technique, device and the computer equipment of vehicle caseload |
| CN110619750A (en) * | 2019-08-15 | 2019-12-27 | 重庆特斯联智慧科技股份有限公司 | Intelligent aerial photography identification method and system for illegal parking vehicle |
| CN112016735A (en) * | 2020-07-17 | 2020-12-01 | 厦门大学 | Patrol route planning method and system based on traffic violation hotspot prediction and readable storage medium |
| CN113269042A (en) * | 2021-04-25 | 2021-08-17 | 安徽银徽科技有限公司 | Intelligent traffic management method and system based on running vehicle violation identification |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| Fang et al. | Attention meets long short-term memory: A deep learning network for traffic flow forecasting | |
| Xiao et al. | How likely am I to find parking?–A practical model-based framework for predicting parking availability | |
| Jeon et al. | Monte Carlo simulation-based traffic speed forecasting using historical big data | |
| Fukuda et al. | Short‐term prediction of traffic flow under incident conditions using graph convolutional recurrent neural network and traffic simulation | |
| CN109360421B (en) | Traffic information prediction method and device based on machine learning and electronic terminal | |
| Li | Traffic incident duration analysis and prediction models based on the survival analysis approach | |
| Adewumi et al. | Application of chaos theory in the prediction of motorised traffic flows on urban networks | |
| Yang et al. | Application of data science technologies in intelligent prediction of traffic congestion | |
| CN112861833B (en) | Vehicle lane level positioning method and device, electronic equipment and computer readable medium | |
| Yang et al. | Toward a real‐time smart parking data management and prediction (SPDMP) system by attributes representation learning | |
| Praveen et al. | RETRACTED ARTICLE: smart traffic management system in metropolitan cities | |
| Zheng et al. | Real-time predication and navigation on traffic congestion model with equilibrium Markov chain | |
| WO2023207411A1 (en) | Traffic determination method and apparatus based on spatio-temporal data, and device and medium | |
| Liu et al. | Modelling the energy consumption of electric vehicles under uncertain and small data conditions | |
| Serok et al. | Identification, cost evaluation, and prioritization of urban traffic congestions and their origin | |
| CN113837635A (en) | Risk detection processing method, device and equipment | |
| Zhao et al. | Highway traffic crash risk prediction method considering temporal correlation characteristics | |
| Li et al. | Conditional inference tree‐based analysis of hazardous traffic conditions for rear‐end and sideswipe collisions with implications for control strategies on freeways | |
| Kasera et al. | Parking slot occupancy prediction using LSTM | |
| Putatunda et al. | Travel time prediction in real time for GPS taxi data streams and its applications to travel safety | |
| Gallo | Models, algorithms, and equilibrium conditions for the simulation of autonomous vehicles in exclusive and mixed traffic | |
| Zhang et al. | Toward large-scale graph-based traffic forecasting: A data-driven network partitioning approach | |
| CN116824868A (en) | Method, device, equipment and medium for identifying illegal parking points and predicting congestion of vehicles | |
| CN114255588A (en) | Regional motor vehicle illegal parking prediction method, storage medium and device | |
| Bing et al. | Short‐Term Traffic Flow Forecasting Method Based on LSSVM Model Optimized by GA‐PSO Hybrid Algorithm |
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 |