A smart city vehicle flow monitoring system based on the Internet of Things

Abstract

The present invention discloses a smart city vehicle flow monitoring system based on the Internet of Things, relates to the technical field of the Internet of Things, and comprises a monitoring center, wherein the monitoring center is connected with a data acquisition module, a data preprocessing module, an intelligent processing module and an interactive analysis module; the vehicle flow video information and vehicle information in the smart city are collected by the data acquisition module; the vehicle information obtained is preprocessed in the data preprocessing module to obtain a standard vehicle atlas; the intelligent processing module processes the obtained vehicle flow video information to obtain a matching characteristic point graph; the interactive analysis module constructs a vehicle flow display graph according to the obtained matching characteristic point graph, and analyzes the vehicle flow display graph to obtain the inspection vehicle capacity; the accuracy and precision of extracting vehicle images are increased, so that users can view the traffic flow situation in real time, reduce traffic congestion and traffic accidents, and improve traffic efficiency.

Description

Technical Field

[0001] The present invention relates to the technical field of Internet of Things, and in particular to a smart city vehicle flow monitoring system based on the Internet of Things. Background Art

[0002] The Internet of Things refers to various physical devices and objects that are connected and interacted through the Internet. These devices and objects are connected through embedded sensors, software and networks, and can collect and exchange data, and realize intelligent automatic operation and monitoring. The core concept of the Internet of Things is to connect sensors, devices and objects to the Internet, so that they can communicate and interact through the network, and remotely monitor, control and exchange data through the Internet to realize data collection, analysis and application.

[0003] Smart city is an urban development concept that comprehensively applies information technology, Internet of Things, big data, artificial intelligence and other advanced technologies. It aims to improve the manageability, sustainability and quality of life of residents in the city through digitalization, intelligence and sustainable development. The goal of smart city is to provide a more efficient, convenient, environmentally friendly and livable urban environment. Through technologies such as the Internet, Internet of Things and sensors, a large amount of urban data is collected and analyzed to achieve information sharing, interconnection and intelligent management of various systems and departments within the city. However, with the acceleration of urbanization, urban traffic problems have become increasingly prominent, and traditional traffic monitoring methods can no longer meet the needs of modern urban traffic management. Therefore, monitoring the traffic flow of smart cities through the Internet of Things has important theoretical and practical significance.

[0004] How to use the Internet of Things to collect and process the traffic video information of the smart city, obtain the matching characteristic point map, construct the traffic display map according to the matching characteristic point map, and analyze the traffic display map to obtain the inspection vehicle capacity; this is the problem we need to solve; for this purpose, a smart city traffic flow monitoring system based on the Internet of Things is now provided. Summary of the invention

[0005] The purpose of the present invention can be achieved through the following technical solutions:

[0006] A smart city vehicle flow monitoring system based on the Internet of Things includes a monitoring center, wherein the monitoring center is connected to a data acquisition module, a data preprocessing module, an intelligent processing module, and an interactive analysis module;

[0007] The data acquisition module is used to collect traffic video information and vehicle information in the smart city;

[0008] The process of the data acquisition module collecting vehicle flow video information includes:

[0009] A smart road map is constructed according to the actual road conditions of the smart city, and an information collection terminal is set according to the actual road conditions of the smart city, wherein the information collection terminal includes a plurality of information capture points, and the set information collection terminal is marked at a corresponding position of the smart road map, and the corresponding position of the information collection terminal in the smart road map is marked as a collection point, and a remote transmission link is constructed between the collection point and the information collection terminal;

[0010] The information collection terminal is used to collect road monitoring information in the smart city, and to capture the collected road monitoring information through the information capture point to obtain traffic video information and video time;

[0011] The obtained traffic video information is uploaded to the corresponding collection point of the information collection terminal through a remote transmission link.

[0012] Screening the obtained vehicle information to obtain vehicle images and vehicle labels;

[0013] The obtained vehicle images are subjected to feature point inspection to obtain a vehicle model edge point map, and a standard vehicle image set is constructed based on the obtained vehicle model edge point map.

[0014] The process of performing feature inspection on the vehicle pattern comprises:

[0015] Standardize the obtained vehicle drawings to obtain standardized vehicle drawings;

[0016] The obtained standard vehicle pattern is edge-captured to obtain the standard vehicle edge, edge identification points are inserted at the standard vehicle edge to obtain a vehicle model edge point map, identity points are set in the vehicle model edge point map, and vehicle labels associated with the standard vehicle pattern are uploaded to the identity points.

[0017] The process of processing the obtained traffic video information includes:

[0018] Perform frame conversion on the obtained traffic flow video information to obtain a single-chip image frame, and obtain the image frame time point corresponding to the single-chip image frame according to the obtained video time;

[0019] Constructing a frame time sequence according to the obtained image frame time points, and performing time sorting on the single image frames according to the constructed frame time sequence to obtain an image frame time sequence;

[0020] Input requirements are set, and according to the obtained input requirements, sizes of individual frames in the frame time sequence are coordinated to obtain standard frames, and the obtained standard frames are marked as input frames.

[0021] Setting a feature filter matrix according to the obtained standard image frame, wherein the feature filter matrix includes a matrix size and a matrix quantity;

[0022] A static extraction layer is set according to the obtained square matrix size and number of square matrices, the obtained feature filter square matrix and the input image frame are uploaded to the static extraction layer, and the obtained feature filter square matrix and the input image frame are subjected to sliding extraction to obtain a feature image frame;

[0023] Perform feature point inspection on the obtained feature map frame to obtain a feature point map, perform property matching between the obtained feature point map and a standard vehicle map set to obtain a matching feature point map, and mark the identity points of the matching feature point map as identification identity points.

[0024] The process of building a traffic flow display diagram includes:

[0025] According to the standard vehicle atlas, the matching characteristic point graphs are grouped into types to obtain type point graphs, and the identification identity points of the type point graphs are marked to obtain identification identity points;

[0026] According to the collected points in the obtained smart road map, the identification points of the matching characteristic point map are marked at the corresponding positions of the smart road map to obtain a traffic display map.

[0027] The process of obtaining inspection vehicle capacity includes:

[0028] A check frame is set in the traffic flow display diagram, and the check frame is provided with an identity sensing point and a time sensing point. The time sensing point is used to record the moving start point and the moving end point of the check frame, and the check duration is obtained according to the obtained moving start point and the moving end point;

[0029] The inspection box is used to move regularly in the traffic flow display diagram, and the interval in which the inspection box moves regularly in the traffic flow display diagram is marked as the moving interval, and the moving interval is matched with the road in the traffic flow display diagram to obtain the inspection distance corresponding to the moving interval;

[0030] A response instruction is issued by the identification identity point within the moving interval, the response instruction is received at the identity sensing point, and the identity sensing points are counted based on the inspection time to obtain the number of types;

[0031] The inspection vehicle capacity is obtained based on the obtained inspection time, number of types and inspection distance.

[0032] Compared with the prior art, the beneficial effects of the present invention are as follows: by preprocessing the collected vehicle information, a vehicle model edge point map is obtained, and a standard vehicle atlas is constructed according to the vehicle model edge point map; the collected vehicle flow video information is processed to obtain a feature map frame, feature point inspection is performed on the feature map frame, and the feature point map is matched with the standard vehicle atlas to obtain a matching feature point map, thereby enhancing the clarity of the video extraction process, facilitating the acquisition of a higher-precision vehicle picture, facilitating the identification of vehicle features, and reducing identification errors;

[0033] The matching feature point graph is grouped by type to obtain a type point graph, and the identification point is marked in the type point graph. A traffic display graph is constructed based on the smart road map. A check frame is constructed by the position of the identification point in the traffic display graph. The inspection vehicle capacity within the moving range is obtained through the check frame. An interactive display graph is obtained for users to view the traffic situation in real time, which helps traffic management departments to improve road conditions and optimize traffic flow, reduce traffic congestion and traffic accidents, and improve traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0034] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings required for use in the embodiments or the description of the prior art will be briefly introduced below. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0035] Figure 1 It is a schematic diagram of the present invention. DETAILED DESCRIPTION

[0036] The technical solution of the present invention will be clearly and completely described below in conjunction with the embodiments. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

[0037] like Figure 1 As shown, a smart city vehicle flow monitoring system based on the Internet of Things includes a monitoring center, and the monitoring center is connected to a data acquisition module, a data preprocessing module, an intelligent processing module, and an interactive analysis module;

[0038] The data acquisition module is used to collect traffic video information and vehicle information in the smart city. The specific process includes:

[0039] Build a smart road map based on the actual road conditions of the smart city, and mark the roads in the smart city at the corresponding positions in the smart road map;

[0040] It should be further explained that, in the specific implementation process, the traffic roads in the constructed smart road map are proportionally reduced according to the proportion of the actual roads, and the motor vehicle lanes, non-motor vehicle lanes and sidewalks on the roads are marked with different colors to distinguish different types of roads;

[0041] Get all vehicle information through smart cities;

[0042] An information collection terminal is set according to the actual road conditions of the smart city obtained, wherein the information collection terminal includes a plurality of information capture points, and the obtained information collection terminal is marked as i, where i=1, 2, 3, ..., v1, and v1 is a positive integer;

[0043] Mark the set information collection terminal at the corresponding position of the smart road map, mark the corresponding position of the information collection terminal in the smart road map as a collection point, and build a remote transmission link from the collection point to the information collection terminal;

[0044] The information collection terminal is used to collect road monitoring information in the smart city, and to capture the collected road monitoring information through the information capture point to obtain traffic video information and video time;

[0045] The obtained traffic video information is uploaded to the corresponding collection point of the information collection terminal through a remote transmission link.

[0046] The data preprocessing module is used to preprocess the obtained vehicle information to obtain a standard vehicle atlas. The specific process includes:

[0047] Screening the obtained vehicle information to obtain vehicle images and vehicle labels;

[0048] Perform feature point inspection on the obtained vehicle pattern to obtain the vehicle model edge point map;

[0049] It should be further explained that, in the specific implementation process, the feature point inspection process includes:

[0050] The obtained vehicle drawings are subjected to graphic standardization to obtain standardized vehicle drawings. Specifically, the graphic standardization is to cut the obtained vehicle drawings into drawings with the same length and width;

[0051] Perform edge capture on the obtained standard vehicle pattern to obtain the standard vehicle edge, insert edge recognition points at the standard vehicle edge, and obtain a vehicle model edge point map;

[0052] The edge recognition points are used for vehicle feature matching. If all edge recognition points can be successfully matched, the vehicle model of the matching object is the matched standard vehicle pattern;

[0053] And set an identity point in the vehicle model edge point map, and upload the vehicle label associated with the standard vehicle pattern to the identity point. In this embodiment, the identity point is set at the license plate position of the vehicle model edge point map;

[0054] A standard vehicle atlas is constructed according to the obtained vehicle model edge point map, and the obtained vehicle model edge point map is uploaded to the standard vehicle atlas;

[0055] It should be further explained that, in the specific implementation process, constructing a standard data set is to record images of existing vehicles, obtain vehicle drawings, and mark the vehicle drawings with vehicle labels. The vehicle labels include vehicle categories, vehicle brands and vehicle models. The vehicle categories include cars, trucks, buses, bicycles, electric vehicles and motorcycles.

[0056] The intelligent processing module is used to process the obtained traffic flow video information to obtain a matching characteristic point map. The specific process includes:

[0057] Perform frame conversion on the obtained traffic flow video information to obtain a single-chip image frame, and obtain the image frame time point corresponding to the single-chip image frame according to the obtained video time;

[0058] Constructing a frame time sequence according to the obtained image frame time points, and performing time sorting on the single image frames according to the constructed frame time sequence to obtain an image frame time sequence;

[0059] Setting input requirements, including length, width, and number of channels;

[0060] It should be further explained that, in the specific implementation process, the input requirement indicates the pixel requirement of the image. For example, the input requirement may be 224x224 or 299x299 pixels. If the resolution of the image is to be clearer, the input requirement may be increased to 600x600 or 800x800 pixels. The number of channels indicates the number of colors of the image. The more channels there are, the clearer the colors of the image are. In this embodiment, the number of channels is 3.

[0061] According to the input requirements obtained, the sizes of the individual image frames in the image frame time sequence are coordinated to obtain standard image frames, and the obtained standard image frames are marked as input image frames;

[0062] Setting a feature filter matrix according to the obtained standard image frame, wherein the feature filter matrix includes a matrix size and a matrix quantity, wherein the matrix size includes a matrix length, a height, and a matrix shape;

[0063] It should be further explained that, in a specific implementation process, the size of the square array determines the pixel range for sliding in the input image frame. For example, the size of the square array is 3×3, 5×5, or 7×7. The shape of the square array includes a circle, a square, or an irregular shape. In this embodiment, the shape of the square array is a square.

[0064] It should be further explained that, in the specific implementation process, the number of square arrays is proportional to the clarity of the image extraction, and under normal circumstances, each feature filter square array is different, which is determined by the square array size. By setting different square array sizes, different feature filter square arrays can be obtained. Therefore, multiple different square array sizes can be set for the same traffic video information, thereby obtaining several feature filter square arrays, and when the feature image is extracted from the input image frame through the feature filter square array, the clarity of the extracted image can be increased;

[0065] A static extraction layer is set according to the obtained square matrix size and number of square matrices, and the obtained feature filter square matrix and input image frame are uploaded to the static extraction layer;

[0066] It should be further explained that, in the specific implementation process, if there are multiple feature filter matrices in the static extraction layer, the feature filter matrices are sorted according to the set time sequence, and the sorted feature filter matrices are respectively subjected to sliding extraction with the input image frames in turn, until all the input image frames have completed sliding extraction with the feature filter matrices, and the remaining feature filter matrices are respectively subjected to sliding extraction with the input image frames in turn; in particular, the matrix size of the feature filter matrix uploaded to the static extraction layer is public, and all input image frames can obtain the matrix size of the feature filter matrix;

[0067] In the set static extraction layer, the obtained feature filter matrix and the input image frame are subjected to sliding extraction to obtain a feature image frame;

[0068] It should be further explained that, in the specific implementation process, the sliding extraction means that in the static extraction layer, any point in the input frame is selected as the starting point, the feature filter matrix starts from the starting point and slides in the input frame, and the dot product of the overlapping part in the interaction process is calculated and recorded as a partial frame, and then the feature filter matrix is ​​slid in the input frame in turn, and the partial frames are recorded until all parts in the input frame complete the dot product with the feature filter matrix, and the obtained partial frames are combined according to the input frame to obtain the feature frame;

[0069] Performing feature point inspection on the obtained feature map frame to obtain a feature point map, performing property matching between the obtained feature point map and a standard vehicle map set to obtain a matching feature point map, and marking the identity points of the matching feature point map as identification identity points;

[0070] It should be further explained that, in the specific implementation process, the property matching is to match the edge identification points in the characteristic point map with the vehicle model edge point map. If all the edge identification points are successfully matched, a matching characteristic point map is obtained, that is, the matching characteristic point map and the vehicle model edge point map are the same model vehicle.

[0071] The interactive analysis module is used to construct a traffic flow display diagram according to the obtained matching characteristic point diagram, and obtain the inspection vehicle capacity by analyzing the traffic flow display diagram. The specific process includes:

[0072] According to the standard vehicle atlas, the matching characteristic point graphs are grouped into types to obtain a first type point graph, a second type point graph, and a third type point graph;

[0073] It should be further explained that, in this embodiment, the first type of point diagram includes cars and motorcycles, the second type of point diagram includes bicycles and electric vehicles, and the third type of point diagram includes trucks and buses;

[0074] Record the identification identity point of the matching characteristic point map in the first type of point map as the first identification identity point, record the identification identity point of the matching characteristic point map in the second type of point map as the second identification identity point, and record the identification identity point of the matching characteristic point map in the third type of point map as the third identification identity point;

[0075] According to the collected points in the obtained smart road map, the identification points of the matching characteristic point map are marked at the corresponding positions of the smart road map to obtain a traffic flow display map;

[0076] It should be further explained that, in the specific implementation process, different types of matching characteristic point maps are marked with different colors to distinguish them. In this embodiment, the first identification point is marked with red, the second identification point is marked with green, and the third identification point is marked with blue;

[0077] Setting a check frame in the traffic flow display diagram, the check frame including a first identity sensing point, a second identity sensing point, a third identity sensing point and a time sensing point;

[0078] It should be further explained that, in the specific implementation process, the size of the inspection frame can be changed and is set according to actual needs. For example, the size of the inspection frame is set to 5×5 or 6×6;

[0079] The inspection frame is used to move regularly in the traffic display map, and the interval in which the inspection frame moves regularly in the traffic display map is marked as the moving interval, and the moving interval is matched with the road in the traffic display map to obtain the inspection distance corresponding to the moving interval, and the obtained inspection distance is marked as L j , where j represents the number of the inspection box, j=1, 2, 3, ..., v2, v2 is a positive integer;

[0080] The time sensing point is used to record the time when the inspection frame starts to move to the moving interval and the time when it leaves the moving interval, obtain the moving start point and the moving end point, mark the obtained moving start point and the moving end point as ta and tb respectively, obtain the inspection duration according to the obtained moving start point and the moving end point, and mark the obtained inspection duration as Δt, where Δt=tb-ta;

[0081] The first identification identity point, the second identification identity point and the third identification identity point in the moving interval send out a first response instruction, a second response instruction and a third response instruction respectively, the first identity sensing point receives the first response instruction, the second identity sensing point receives the second response instruction and the third identity sensing point receives the third response instruction, and based on the inspection time, the identity sensing points are counted respectively to obtain the first type number, the second type number and the third type number, and are marked as S1, S2 and S3;

[0082] The inspection vehicle capacity is obtained according to the obtained inspection time, the number of the first type, the number of the second type, the number of the third type and the inspection distance, and the obtained inspection vehicle capacity is marked as CR j ,in t1, t2 and t3 represent the time required for the first type of dot graph, the second type of dot graph and the third type of dot graph to pass through the moving interval, respectively. β1, β2 and β3 are proportional factors. In particular, when only the inspection vehicle capacity of the first type of dot graph needs to be counted, β2=β3=0. If the inspection vehicle capacity of the second type of dot graph is counted, β1=β3=0. If the inspection vehicle capacity of the third type of dot graph is counted, β2=β1=0.

[0083] The estimated time is obtained based on the obtained inspection capacity, and the estimated time is marked as T 预 ,in v 平 Indicates the average speed of the inspection route;

[0084] It should be further explained that, in the specific implementation process, by marking the vehicle position on the traffic display map, the user can intuitively observe the traffic conditions of the driving section, and by selecting the required inspection box, the user can obtain the traffic conditions of the moving section in real time, and then select the road suitable for the user in combination with the expected travel time, and the vehicle types are distinguished by different colors, so that the user can pay attention to the traffic conditions of the road where the vehicle is located. For example, when the user passes the inspection section by electric vehicle, the vehicle conditions in the inspection box and the expected travel time can be checked in advance; at the same time, the inspection vehicle capacity of different lanes can be uploaded to the road management department, and the road can be optimized and improved, and the road can be widened or the route can be modified to improve the user's driving experience, shorten the driving time, and increase the safety of the journey;

[0085] In particular, if the size of the inspection box is the same as the size of the traffic flow display map, it means that the entire traffic flow display map is inspected, observation points are set according to the road conditions in the traffic flow display map, an observation range is set for the observation points, and the observation time of the observation points is recorded. The number of first identification points, second identification points and third identification points within the observation range are counted based on the observation points, the obtained numbers are marked and associated with the corresponding observation time, the number of different types of vehicles within the observation range is displayed through the observation points, and the inspection vehicle capacity of each observation point is obtained.

[0086] The preferred embodiments of the present invention disclosed above are only used to help explain the present invention. The preferred embodiments do not describe all the details in detail, nor do they limit the invention to only specific implementation methods. Obviously, many modifications and changes can be made according to the content of this specification. This specification selects and specifically describes these embodiments in order to better explain the principles and practical applications of the present invention, so that those skilled in the art can understand and use the present invention well. The present invention is limited only by the claims and their full scope and equivalents.

Claims

1. A smart city vehicle flow monitoring system based on the Internet of Things, including a monitoring center, characterized in that: The monitoring center is connected to a data acquisition module, a data preprocessing module, an intelligent processing module and an interactive analysis module; The data acquisition module is used to collect traffic video information and vehicle information in the smart city; The data preprocessing module is used to preprocess the obtained vehicle information to obtain a standard vehicle atlas; Screening the obtained vehicle information to obtain vehicle images and vehicle labels; Perform feature point inspection on the obtained vehicle images to obtain a vehicle model edge point map, and construct a standard vehicle atlas based on the obtained vehicle model edge point map; Standardize the obtained vehicle drawings to obtain standardized vehicle drawings; Perform edge capture on the obtained standard vehicle pattern to obtain the standard vehicle edge, insert edge identification points at the standard vehicle edge, obtain a vehicle model edge point map, set identity points in the vehicle model edge point map, and upload the vehicle label associated with the standard vehicle pattern to the identity point; The intelligent processing module is used to process the obtained vehicle flow video information to obtain a matching characteristic point diagram; Perform frame conversion on the obtained traffic flow video information to obtain a single-chip image frame, and obtain the image frame time point corresponding to the single-chip image frame according to the obtained video time; Constructing a frame time sequence according to the obtained image frame time points, and performing time sorting on the single image frames according to the constructed frame time sequence to obtain an image frame time sequence; Setting input requirements, and according to the obtained input requirements, respectively coordinating the sizes of the individual image frames in the image frame time sequence to obtain standard image frames, and marking the obtained standard image frames as input image frames; Setting a feature filter matrix according to the obtained standard image frame, wherein the feature filter matrix includes a matrix size and a matrix quantity; A static extraction layer is set according to the obtained square matrix size and number of square matrices, the obtained feature filter square matrix and the input image frame are uploaded to the static extraction layer, and the obtained feature filter square matrix and the input image frame are subjected to sliding extraction to obtain a feature image frame; Performing feature point inspection on the obtained feature map frame to obtain a feature point map, performing property matching between the obtained feature point map and a standard vehicle map set to obtain a matching feature point map, and marking the identity points of the matching feature point map as identification identity points; The interactive analysis module is used to construct a traffic flow display diagram according to the obtained matching characteristic point diagram, and analyze the traffic flow display diagram to obtain the inspection vehicle capacity; According to the standard vehicle atlas, the matching characteristic point graphs are grouped into types to obtain type point graphs, and the identification identity points of the type point graphs are marked to obtain identification identity points; According to the collected points in the obtained smart road map, the identification points of the matching characteristic point map are marked at the corresponding positions of the smart road map to obtain a traffic flow display map; A check frame is set in the traffic flow display diagram, and the check frame is provided with an identity sensing point and a time sensing point. The time sensing point is used to record the moving start point and the moving end point of the check frame, and the check duration is obtained according to the obtained moving start point and the moving end point; The inspection box is used to move regularly in the traffic flow display diagram, and the interval in which the inspection box moves regularly in the traffic flow display diagram is marked as a moving interval, and the moving interval is matched with the road in the traffic flow display diagram to obtain the inspection distance corresponding to the moving interval; A response instruction is issued by the identification identity point within the moving interval, the response instruction is received at the identity sensing point, and the identity sensing points are counted based on the inspection time to obtain the number of types; The inspection vehicle capacity is obtained based on the obtained inspection time, number of types and inspection distance.

2. According to the Internet of Things-based smart city vehicle flow monitoring system of claim 1, it is characterized in that: The process of the data acquisition module collecting vehicle flow video information includes: A smart road map is constructed according to the actual road conditions of the smart city, and an information collection terminal is set according to the actual road conditions of the smart city, wherein the information collection terminal includes a plurality of information capture points, and the set information collection terminal is marked at a corresponding position of the smart road map, and the corresponding position of the information collection terminal in the smart road map is marked as a collection point, and a remote transmission link is constructed between the collection point and the information collection terminal; The information collection terminal is used to collect road monitoring information in the smart city, and to capture the collected road monitoring information through the information capture point to obtain traffic video information and video time; The obtained traffic video information is uploaded to the corresponding collection point of the information collection terminal through a remote transmission link.

Images