Traffic data visualization method and system based on Internet of Things

Abstract

The invention provides a traffic data visualization method and system based on the Internet of Things, and relates to the technical field of traffic, and the method comprises the steps: collecting the traffic data of a target intersection based on the Internet of Things; image segmentation processing is carried out on the traffic data, an image segmentation result is obtained, and the image segmentation result comprises a plurality of vehicle images; according to the image segmentation result, analyzing and obtaining vehicle number data of the target intersection, and according to the vehicle data number, analyzing and obtaining first congestion degree information of the target intersection; according to an image segmentation result, identifying and obtaining a plurality of vehicle head distances of a plurality of vehicles in the target intersection, and calculating and obtaining second congestion degree information; and based on the first congestion degree information and the second congestion degree information, carrying out congestion identification and classification on the target intersection to obtain a comprehensive congestion degree and corresponding congestion identification information, and carrying out visual identification on the target intersection. According to the invention, the technical effect of traffic data visualization processing is achieved.

Description

Technical Field

[0001] The present invention relates to the field of traffic technology, and in particular to a traffic data visualization method and system based on the Internet of Things. Background Art

[0002] With the development of Internet of Things technology, everything is connected, and there are more and more applications of software combined with the network to empower things. Among them, the collection, monitoring and analysis of traffic data through the Internet of Things can effectively improve the efficiency and effectiveness of traffic management.

[0003] However, the current level of integration between the Internet of Things and traffic management is relatively shallow, and can only perform relatively simple image classification and recognition, but is unable to perform visual traffic data processing. Summary of the invention

[0004] The present application provides a traffic data visualization method and system based on the Internet of Things, which is used to solve the technical problem of traffic data processing that cannot be visualized in the prior art.

[0005] The first aspect of the present application provides a traffic data visualization method based on the Internet of Things, the method comprising:

[0006] Based on the Internet of Things, traffic data of a target intersection is collected, wherein the traffic data includes a traffic image within the target intersection;

[0007] Performing image segmentation processing on the traffic data to obtain an image segmentation result, wherein the image segmentation result includes a plurality of vehicle images;

[0008] According to the image segmentation result, the vehicle quantity data of the target intersection is analyzed and obtained, and according to the vehicle quantity data, the first congestion degree information of the target intersection is analyzed and obtained;

[0009] According to the image segmentation result, multiple headway distances of multiple vehicles in the target intersection are identified and obtained, and second congestion degree information is calculated;

[0010] Based on the first congestion degree information and the second congestion degree information, congestion identification and classification of the target intersection are performed to obtain a comprehensive congestion degree and corresponding congestion identification information, and the target intersection is visually identified.

[0011] The second aspect of the present application provides a traffic data visualization system based on the Internet of Things, the system comprising:

[0012] A traffic image acquisition module, used to collect traffic data of a target intersection based on the Internet of Things, wherein the traffic data includes a traffic image within the target intersection;

[0013] A vehicle image segmentation module, used to perform image segmentation processing on the traffic data to obtain an image segmentation result, wherein the image segmentation result includes a plurality of vehicle images;

[0014] A first congestion analysis module is used to analyze and obtain vehicle quantity data at a target intersection according to the image segmentation result, and to analyze and obtain first congestion degree information of the target intersection according to the vehicle quantity data;

[0015] A second congestion analysis module is used to identify and obtain multiple headway distances between multiple vehicles in the target intersection according to the image segmentation result, and calculate and obtain second congestion degree information;

[0016] The visualization identification module is used to identify and classify the congestion of the target intersection based on the first congestion information and the second congestion information, obtain the comprehensive congestion degree and the corresponding congestion identification information, and perform visualization identification on the target intersection.

[0017] According to a third aspect of the present application, a computer device is provided, comprising a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method in the first aspect are implemented.

[0018] According to a fourth aspect of the present application, a computer-readable storage medium is provided, wherein the computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method in the first aspect are implemented.

[0019] One or more technical solutions provided in this application have at least the following technical effects or advantages:

[0020] The technical solution provided by the present application collects traffic data of a target intersection, performs image segmentation, obtains multiple vehicle images, identifies the number of vehicles, obtains first congestion information, analyzes the headway between multiple vehicles in the target intersection, obtains second congestion information of the target intersection, further comprehensively analyzes the comprehensive congestion of the target intersection, determines congestion identification information for marking the intersection based on the information, performs identification, realizes visualization processing and identification of traffic data at the intersection, improves the visualization level of traffic data processing, makes it more direct and concrete, and thus improves the efficiency and effectiveness of traffic management, while avoiding the problem of traffic management effectiveness being affected by the degree of visualization of traffic data, thereby achieving visualization processing of traffic data based on the Internet of Things, improving the visualization level of traffic data and the degree of accuracy of description, improving the efficiency of traffic data processing, and lowering the threshold for understanding and processing of traffic management. BRIEF DESCRIPTION OF THE DRAWINGS

[0021] Figure 1 A flow chart of a traffic data visualization method based on the Internet of Things provided in this application;

[0022] Figure 2 A schematic diagram of the structure of a traffic data visualization system based on the Internet of Things provided in this application;

[0023] Figure 3 A schematic diagram of the structure of an exemplary computer device according to an embodiment of the present application.

[0024] Explanation of the reference numerals: traffic image acquisition module 11 , vehicle image segmentation module 12 , first congestion analysis module 13 , second congestion analysis module 14 , visual identification module 15 , computer device 300 , memory 301 , processor 302 , communication interface 303 , bus architecture 304 . DETAILED DESCRIPTION

[0025] The present application provides a traffic data visualization method and system based on the Internet of Things, which is used to solve the technical problem of traffic data processing that cannot be visualized in the prior art.

[0026] Embodiment 1

[0027] like Figure 1 As shown, the present application provides a traffic data visualization method based on the Internet of Things, the method comprising:

[0028] S100: Based on the Internet of Things, traffic data of a target intersection is collected, where the traffic data includes a traffic image in the target intersection;

[0029] In an embodiment of the present application, based on the Internet of Things technology, traffic data of a target intersection to be subjected to traffic data visualization processing is collected, and the traffic data includes a traffic image within the target intersection.

[0030] The target intersection may be an intersection on any road, such as any crossroads on a city road.

[0031] Based on the Internet of Things technology, for example, an image acquisition device connected to a network deployed in the target intersection is used to collect images in the target intersection as traffic images, and the traffic images include roads and vehicles in the target intersection.

[0032] S200: performing image segmentation processing on the traffic data to obtain an image segmentation result, wherein the image segmentation result includes a plurality of vehicle images;

[0033] In an embodiment of the present application, image segmentation processing is performed on the traffic image of the target intersection of the traffic data, and the vehicle images of the vehicles in the target intersection are segmented and identified, and the image segmentation results are obtained as a data basis for analyzing the congestion level of the target intersection according to the vehicles.

[0034] Step S200 in the method provided in the embodiment of the present application includes:

[0035] S210: Train traffic image segmenter based on semantic segmentation;

[0036] S220: Using the traffic image segmentor to segment the traffic image to obtain the image segmentation result.

[0037] In an embodiment of the present application, a traffic image segmentor for segmenting traffic images is trained based on semantic segmentation to perform automatic intelligent segmentation of images of multiple vehicles in a traffic image, thereby improving the efficiency and accuracy of traffic data visualization processing.

[0038] Step S210 in the method provided in the embodiment of the present application includes:

[0039] Based on semantic segmentation, build encoder and decoder;

[0040] According to the traffic monitoring data of the target intersection, a sample traffic data set is obtained;

[0041] Segment and mark the vehicle images in the sample traffic data to obtain a sample image segmentation result set;

[0042] Using the sample traffic data set and the sample image segmentation result set, supervised training is performed on the encoder and the decoder until convergence;

[0043] Based on the encoder and decoder trained to convergence, a traffic image segmentor is obtained.

[0044] In the embodiment of the present application, the traffic image segmentor is constructed and trained based on semantic segmentation in machine learning.

[0045] According to the traffic monitoring data of the target intersection in the past, the traffic images monitored and acquired in the past are extracted to obtain the sample traffic data set. For example, the monitoring data of the target intersection in the past month stored in the server can be retrieved as the sample traffic data set.

[0046] Furthermore, based on manual labeling, the vehicle images in each sample traffic image in the sample traffic data set are divided and labeled. Specifically, the edges of the vehicle images are divided as boundaries and labeled with the vehicle labeling information. In this way, the division and labeling of the vehicle images in the sample traffic data are completed, and a set of sample image segmentation results is obtained.

[0047] Based on the semantic segmentation algorithm in machine learning, an encoder and a decoder are constructed. The encoder and the decoder include a convolutional layer, a pooling layer, a fully connected layer, and a deconvolutional layer, a depooling layer, and a fully connected layer, respectively. The features in the input image can be extracted, and the vehicle can be recognized and identified based on the features, and the feature image of the segmented vehicle image, i.e., the image segmentation result, can be output.

[0048] Using the sample traffic data set and the sample image segmentation result set, with the traffic data as input and the image segmentation result as output, the constructed encoder and decoder are supervised trained, and the network parameters are adjusted during the training process until the convergence requirements are met.

[0049] For example, the convergence requirement may be that the accuracy rate reaches 85%, or converges with respect to a sample traffic data set and a sample image segmentation result set.

[0050] Based on the encoder and decoder trained to convergence, the traffic image segmentor is obtained.

[0051] The trained traffic image segmentor is used to perform image semantic segmentation processing on the traffic image, and vehicle images therein are identified and marked to obtain an image segmentation result, which includes vehicle images of multiple vehicles.

[0052] The embodiment of the present application trains a traffic image segmentor based on semantic segmentation, segments traffic images, and obtains vehicle images, which can improve the accuracy of vehicle-based analysis of intersection congestion and thus lay a foundation for visualization of traffic data.

[0053] S300: Analyze and obtain vehicle quantity data of the target intersection according to the image segmentation result, and analyze and obtain first congestion degree information of the target intersection according to the vehicle quantity data;

[0054] In the embodiment of the present application, based on the number of vehicles in the image segmentation result, a first congestion degree in the target intersection is analyzed and obtained.

[0055] Step S300 in the method provided in the embodiment of the present application includes:

[0056] S310: Counting the number of vehicle images in the image segmentation result to obtain vehicle quantity information;

[0057] S320: Using the vehicle quantity information as first congestion degree information.

[0058] In the embodiment of the present application, based on the image segmentation result, the number of vehicle images therein is counted, and then the number of vehicles in the current target intersection is obtained as vehicle quantity information.

[0059] The vehicle quantity information is used as the current first congestion degree information of the target intersection, reflecting the current congestion degree of the target intersection from the dimension of vehicle quantity, and further serves as the basis for analyzing the comprehensive congestion degree of the target intersection and performing visualization processing.

[0060] S400: identifying and obtaining a plurality of headway distances between a plurality of vehicles in the target intersection according to the image segmentation result, and calculating and obtaining second congestion degree information;

[0061] In an embodiment of the present application, based on multiple vehicle images in the image segmentation result, the headway between adjacent vehicles is analyzed as basic data for another dimension of analyzing the degree of traffic congestion at the intersection, and the second congestion degree information of the target intersection is analyzed and calculated.

[0062] Step S400 in the method provided in the embodiment of the present application includes:

[0063] S410: training and acquiring a headway distance identifier;

[0064] S420: using the vehicle head distance identifier to perform vehicle head distance identification on the multiple vehicle images in the image segmentation result to obtain multiple vehicle head distances;

[0065] S430: Calculate the average of multiple headway distances to obtain the second congestion degree information.

[0066] A headway distance identifier for identifying headway distances is trained, and headway distances are further identified for multiple vehicle images in the image segmentation result.

[0067] Step S410 in the method provided in the embodiment of the present application includes:

[0068] According to the traffic monitoring data in the plurality of intersections, a plurality of sample binary vehicle images are processed and acquired, each sample binary vehicle image including two vehicle images of two adjacent vehicles in front and behind;

[0069] The test obtains the headway distances in multiple sample binary vehicle images to obtain multiple sample headway distances;

[0070] Based on the convolutional neural network, a headway recognition device is constructed;

[0071] Using multiple sample binary vehicle images and multiple sample headway distances, the headway distance identifier is trained until convergence;

[0072] Combining the multiple vehicle images in the image segmentation result to obtain multiple binary vehicle images;

[0073] The vehicle head distance identifier is used to identify the multiple binary vehicle images to obtain multiple vehicle head distances.

[0074] In an embodiment of the present application, based on monitoring data of multiple intersections in a traffic system, a set of sample traffic images in multiple intersections is obtained, and each sample traffic image includes a traffic monitoring image in a certain intersection at a certain point in time.

[0075] Based on the steps of the method in the foregoing content, the vehicle images in the plurality of sample traffic images may be segmented to obtain a set of sample image segmentation results.

[0076] Furthermore, based on the manual annotation method, the two vehicles in front and behind of each sample image segmentation result in the same lane are divided, so that multiple groups of vehicle images of two vehicles distributed adjacent to each other can be obtained in each sample image segmentation result, and the vehicle images of the two vehicles are framed and cropped as sample binary vehicle images, thereby obtaining multiple sample binary vehicle images.

[0077] According to the distance between the fronts of two vehicle images in each sample binary vehicle image and the vehicle size ratio, the front distance between two vehicles in multiple sample binary vehicle images can be calculated, thereby obtaining multiple sample front distances.

[0078] Based on the convolutional neural network, the basic network architecture of the vehicle headway distance identifier is constructed, which includes multiple convolution layers, pooling layers, activation functions and fully connected layers. The features in the image are extracted based on the 3*3 convolution kernel, and the vehicle headway distance is identified.

[0079] The plurality of sample binary vehicle images and the plurality of sample headway distances are used as training data to perform supervised training on the headway distance identifier, and the network parameters are adjusted based on the output headway distances and the loss of the sample headway distances until the convergence condition is met. For example, the convergence condition may be that the accuracy of the headway distance identifier reaches 85%.

[0080] Multiple groups of two adjacent vehicle images in the same lane in the current image segmentation result are framed and cropped to obtain the current multiple binary vehicle images.

[0081] The vehicle head distance identifier is used to identify the current multiple binary vehicle images to obtain the vehicle head distances, so that multiple vehicle head distances can be obtained.

[0082] Furthermore, an average of multiple headway distances is calculated as the second congestion degree information.

[0083] The embodiment of the present application improves the accuracy and comprehensiveness of the analysis by selecting feature data of another dimension of the headway as the data basis for analyzing the degree of congestion at traffic intersections, and improves the efficiency and accuracy of headway recognition by image segmentation and combined recognition of binary vehicle images.

[0084] S500: Based on the first congestion degree information and the second congestion degree information, perform congestion identification and classification of the target intersection, obtain a comprehensive congestion degree and corresponding congestion identification information, and visually identify the target intersection.

[0085] In the embodiment of the present application, based on the first congestion information and the second congestion information analyzed and calculated in the aforementioned content, a more comprehensive congestion identification and classification of the congestion level in the current target intersection is performed, and visual identification is performed to achieve accurate visual processing of traffic data.

[0086] Step S500 in the method provided in the embodiment of the present application includes:

[0087] S510: According to the traffic data monitoring record of the target intersection, a sample first congestion degree information set and a sample second congestion degree information set are processed and obtained;

[0088] S520: Evaluate and obtain a sample comprehensive congestion degree set according to the sample first congestion degree information set and the sample second congestion degree information set;

[0089] S530: Acquire multiple sample congestion identification information, and construct a mapping relationship with the sample comprehensive congestion degree set;

[0090] S540: constructing a comprehensive congestion degree classifier based on a decision tree by using the sample first congestion degree information set, the sample second congestion degree information set and the sample comprehensive congestion degree set;

[0091] S550: Perform congestion degree identification and classification according to the first congestion degree information and the second congestion degree information to obtain a comprehensive congestion degree, and map and obtain the congestion identification information.

[0092] In the embodiment of the present application, according to the traffic data monitoring record of the target intersection, based on the steps of the method in the aforementioned content, the same processing is performed to obtain a sample first congestion information set and a sample second congestion information set.

[0093] Technical personnel in the field of traffic management evaluate and analyze the congestion level of intersections based on the corresponding sample first congestion information and sample second congestion information in the sample first congestion information set and the sample second congestion information set, and obtain a sample comprehensive congestion set.

[0094] Furthermore, based on the visualized intersection congestion signs, multiple sample congestion sign information is obtained. Exemplarily, the multiple sample congestion sign information includes color signs, including red, orange, yellow, green, etc., which respectively correspond to gradually decreasing congestion levels.

[0095] A mapping relationship between multiple sample congestion identification information and a sample comprehensive congestion degree set is constructed, wherein, for example, based on the size distribution of multiple sample comprehensive congestion degrees in the sample comprehensive congestion degree set, multiple intervals can be equally spaced to construct a mapping relationship with multiple sample congestion identification information.

[0096] Furthermore, based on the decision tree algorithm, the sample first congestion information set, the sample second congestion information set and the sample comprehensive congestion set are used to construct a comprehensive congestion classifier, wherein the decision input of the comprehensive congestion classifier is the first congestion information and the second congestion information, and the decision output is the comprehensive congestion.

[0097] The comprehensive congestion degree classifier includes multiple decision nodes, each of which includes a decision judgment threshold. It makes a judgment on the input first congestion degree information or the second congestion degree information, and inputs the judgment result into the next decision node for further judgment. After multiple layers of judgment analysis, the corresponding comprehensive congestion degree can be obtained based on the input first congestion degree information and the second congestion degree information, thereby completing the decision classification of the comprehensive congestion degree.

[0098] The comprehensive congestion degree classifier is used to make decision classification on the first congestion degree information and the second congestion degree information obtained by the current analysis, and the comprehensive congestion degree of the target intersection is obtained, thereby achieving accurate and efficient intersection congestion degree analysis.

[0099] Based on the comprehensive congestion degree, corresponding congestion identification information is mapped and obtained, and the target intersection is visually identified, for example, in an electronic map, to achieve comprehensive and accurate visualization of traffic data.

[0100] In summary, the embodiments of the present application have at least the following technical effects:

[0101] The embodiment of the present application collects traffic data of a target intersection, performs image segmentation, obtains multiple vehicle images, identifies the number of vehicles, obtains first congestion information, analyzes the headway between multiple vehicles in the target intersection, obtains second congestion information of the target intersection, further comprehensively analyzes the comprehensive congestion of the target intersection, determines congestion identification information for marking the intersection based on the information, performs identification, and implements visualization processing and identification of traffic data at the intersection, thereby improving the visualization level of traffic data processing, making it more direct and concrete, and thereby improving the efficiency and effectiveness of traffic management, while avoiding the problem of traffic management effectiveness being affected by the degree of visualization of traffic data, thereby achieving visualization processing of traffic data based on the Internet of Things, improving the visualization level of traffic data and the degree of accuracy of description, improving the efficiency of traffic data processing, and lowering the threshold for understanding and processing of traffic management.

[0102] Embodiment 2

[0103] Based on the same inventive concept as the traffic data visualization method based on the Internet of Things in the aforementioned embodiment, Figure 2 As shown, the present application provides a traffic data visualization system based on the Internet of Things. The specific description of a traffic data visualization method based on the Internet of Things in Embodiment 1 is also applicable to the traffic data visualization system based on the Internet of Things, wherein the system includes:

[0104] A traffic image acquisition module 11 is used to collect traffic data of a target intersection based on the Internet of Things, wherein the traffic data includes a traffic image within the target intersection;

[0105] The vehicle image segmentation module 12 is used to perform image segmentation processing on the traffic data to obtain an image segmentation result, wherein the image segmentation result includes a plurality of vehicle images;

[0106] The first congestion analysis module 13 is used to analyze and obtain vehicle quantity data at the target intersection according to the image segmentation result, and to analyze and obtain first congestion degree information of the target intersection according to the vehicle quantity data;

[0107] The second congestion analysis module 14 is used to identify and obtain multiple headway distances between multiple vehicles in the target intersection according to the image segmentation result, and calculate and obtain second congestion degree information;

[0108] The visual identification module 15 is used to identify and classify the congestion of the target intersection based on the first congestion information and the second congestion information, obtain the comprehensive congestion degree and the corresponding congestion identification information, and visually identify the target intersection.

[0109] Furthermore, the vehicle image segmentation module 12 is also used to implement the following functions:

[0110] Based on semantic segmentation, train traffic image segmentor;

[0111] The traffic image is segmented using the traffic image segmentor to obtain the image segmentation result.

[0112] Among them, based on semantic segmentation, training traffic image segmentation includes:

[0113] Based on semantic segmentation, build encoder and decoder;

[0114] According to the traffic monitoring data of the target intersection, a sample traffic data set is obtained;

[0115] Segment and mark the vehicle images in the sample traffic data to obtain a sample image segmentation result set;

[0116] Using the sample traffic data set and the sample image segmentation result set, supervised training is performed on the encoder and the decoder until convergence;

[0117] Based on the encoder and decoder trained to convergence, a traffic image segmentor is obtained.

[0118] Furthermore, the first congestion analysis module 13 is also used to implement the following functions:

[0119] Counting the number of vehicle images in the image segmentation result to obtain vehicle quantity information;

[0120] The vehicle quantity information is used as first congestion degree information.

[0121] Furthermore, the second congestion analysis module 14 is also used to implement the following functions:

[0122] Train and obtain the headway distance identifier;

[0123] Using the vehicle head distance identifier, performing vehicle head distance identification on a plurality of vehicle images in the image segmentation result to obtain a plurality of vehicle head distances;

[0124] Calculate the average of multiple headway distances to obtain the second congestion degree information.

[0125] Among them, training to obtain the headway distance identifier includes:

[0126] According to the traffic monitoring data in the plurality of intersections, a plurality of sample binary vehicle images are processed and acquired, each sample binary vehicle image including two vehicle images of two adjacent vehicles in front and behind;

[0127] The test obtains the headway distances in multiple sample binary vehicle images to obtain multiple sample headway distances;

[0128] Based on the convolutional neural network, a headway recognition device is constructed;

[0129] Using multiple sample binary vehicle images and multiple sample headway distances, the headway distance identifier is trained until convergence;

[0130] Combining the multiple vehicle images in the image segmentation result to obtain multiple binary vehicle images;

[0131] The vehicle head distance identifier is used to identify the multiple binary vehicle images to obtain multiple vehicle head distances.

[0132] Furthermore, the visual identification module 15 is also used to implement the following functions:

[0133] According to the traffic data monitoring record of the target intersection, a sample first congestion degree information set and a sample second congestion degree information set are processed and obtained;

[0134] Evaluate and obtain a sample comprehensive congestion degree set according to the sample first congestion degree information set and the sample second congestion degree information set;

[0135] Acquire multiple sample congestion identification information, and construct a mapping relationship with the sample comprehensive congestion degree set;

[0136] Using the sample first congestion degree information set, the sample second congestion degree information set and the sample comprehensive congestion degree set, based on a decision tree, construct a comprehensive congestion degree classifier;

[0137] Congestion degree identification and classification are performed according to the first congestion degree information and the second congestion degree information to obtain a comprehensive congestion degree, and the congestion identification information is mapped and obtained.

[0138] Embodiment 3

[0139] like Figure 3 As shown, based on the same inventive concept as a traffic data visualization method based on the Internet of Things in the aforementioned embodiment, the present application also provides a computer device 300, wherein the computer device 300 includes a memory 301 and a processor 302, wherein a computer program is stored in the memory 301, and when the computer program is executed by the processor 302, the steps of a method in an embodiment are implemented.

[0140] The computer device 300 includes: a processor 302, a communication interface 303, and a memory 301. Optionally, the computer device 300 may also include a bus architecture 304. The communication interface 303, the processor 302, and the memory 301 may be interconnected via the bus architecture 304; the bus architecture 304 may be a peripheral component interconnect (PCI) bus or an extended industry standard architecture (EISA) bus. The bus architecture 304 may be divided into an address bus, a data bus, a control bus, etc. For ease of representation, Figure 3 Only one thick line is used in the diagram, but this does not mean that there is only one bus or only one type of bus.

[0141] The processor 302 may be a CPU, a microprocessor, an ASIC, or one or more integrated circuits for controlling the execution of the program of the present application.

[0142] The communication interface 303 uses any transceiver-like device for communicating with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), wired access networks, etc.

[0143] The memory 301 may be a ROM or other types of static storage devices that can store static information and instructions, a RAM or other types of dynamic storage devices that can store information and instructions, or an electrically erasable programmable read only memory (EEPROM), a compact disc (CD ROM) or other optical disc storage, an optical disc storage (including a compressed optical disc, a laser disc, an optical disc, a digital versatile disc, a Blu-ray disc, etc.), a magnetic disk storage medium or other magnetic storage device, or any other medium that can be used to carry or store the desired program code in the form of an instruction or data structure and can be accessed by a computer, but is not limited thereto. The memory may exist independently and be connected to the processor via a bus architecture 304. The memory may also be integrated with the processor.

[0144] The memory 301 is used to store computer-executable instructions for executing the solution of the present application, and the execution is controlled by the processor 302. The processor 302 is used to execute the computer-executable instructions stored in the memory 301, thereby realizing a traffic data visualization method based on the Internet of Things provided in the above embodiment of the present application.

[0145] Embodiment 4

[0146] Based on the same inventive concept as the traffic data visualization method based on the Internet of Things in the aforementioned embodiment, the present application also provides a computer-readable storage medium, which stores a computer program. When the computer program is executed by a processor, the steps of the method in embodiment one are implemented.

[0147] This specification and drawings are merely exemplary illustrations of the present application and are deemed to cover any and all modifications, variations, combinations or equivalents within the scope of the present application. Obviously, a person skilled in the art may make various modifications and variations to the present application without departing from the scope of the present application. Thus, if these modifications and variations of the present application fall within the scope of the present application and its equivalents, the present application intends to include these modifications and variations.

Claims

1. A traffic data visualization method based on the Internet of Things, characterized in that: The method comprises: Based on the Internet of Things, traffic data of a target intersection is collected, wherein the traffic data includes a traffic image within the target intersection; Performing image segmentation processing on the traffic data to obtain an image segmentation result, wherein the image segmentation result includes a plurality of vehicle images; According to the image segmentation result, the vehicle quantity data of the target intersection is analyzed and obtained, and according to the vehicle quantity data, the first congestion degree information of the target intersection is analyzed and obtained; According to the image segmentation result, multiple headway distances of multiple vehicles in the target intersection are identified and obtained, and second congestion degree information is calculated; Based on the first congestion degree information and the second congestion degree information, congestion identification and classification of the target intersection are performed to obtain a comprehensive congestion degree and corresponding congestion identification information, and the target intersection is visually identified.

2. The method according to claim 1, characterized in that: Performing image segmentation processing on the traffic data, the method comprising: Based on semantic segmentation, train traffic image segmentor; The traffic image is segmented using the traffic image segmentor to obtain the image segmentation result.

3. The method according to claim 2, characterized in that Based on semantic segmentation, train the traffic image segmenter, including: Based on semantic segmentation, build encoder and decoder; According to the traffic monitoring data of the target intersection, a sample traffic data set is obtained; Segment and mark the vehicle images in the sample traffic data to obtain a sample image segmentation result set; Using the sample traffic data set and the sample image segmentation result set, supervised training is performed on the encoder and the decoder until convergence; Based on the encoder and decoder trained to convergence, a traffic image segmentor is obtained.

4. The method according to claim 1, characterized in that The method comprises: Counting the number of vehicle images in the image segmentation result to obtain vehicle quantity information; The vehicle quantity information is used as first congestion degree information.

5. The method according to claim 1, characterized in that: The method comprises: Train and obtain the headway distance identifier; Using the vehicle head distance identifier, performing vehicle head distance identification on a plurality of vehicle images in the image segmentation result to obtain a plurality of vehicle head distances; Calculate the average of multiple headway distances to obtain the second congestion degree information.

6. The method according to claim 5, characterized in that Train and obtain the headway distance identifier, including: According to the traffic monitoring data in the plurality of intersections, a plurality of sample binary vehicle images are processed and acquired, each sample binary vehicle image including two vehicle images of two adjacent vehicles in front and behind; The test obtains the headway distances in multiple sample binary vehicle images to obtain multiple sample headway distances; Based on the convolutional neural network, a headway recognition device is constructed; Using multiple sample binary vehicle images and multiple sample headway distances, the headway distance identifier is trained until convergence; Combining the multiple vehicle images in the image segmentation result to obtain multiple binary vehicle images; The vehicle head distance identifier is used to identify the multiple binary vehicle images to obtain multiple vehicle head distances.

7. The method according to claim 1, characterized in that The method comprises: According to the traffic data monitoring record of the target intersection, a sample first congestion degree information set and a sample second congestion degree information set are processed and obtained; Evaluate and obtain a sample comprehensive congestion degree set according to the sample first congestion degree information set and the sample second congestion degree information set; Acquire multiple sample congestion identification information, and construct a mapping relationship with the sample comprehensive congestion degree set; Using the sample first congestion degree information set, the sample second congestion degree information set and the sample comprehensive congestion degree set, based on a decision tree, construct a comprehensive congestion degree classifier; Congestion degree identification and classification are performed according to the first congestion degree information and the second congestion degree information to obtain a comprehensive congestion degree, and the congestion identification information is mapped and obtained.

8. A traffic data visualization system based on the Internet of Things, characterized in that: The system comprises: A traffic image acquisition module, used to collect traffic data of a target intersection based on the Internet of Things, wherein the traffic data includes a traffic image within the target intersection; A vehicle image segmentation module, used to perform image segmentation processing on the traffic data to obtain an image segmentation result, wherein the image segmentation result includes a plurality of vehicle images; A first congestion analysis module is used to analyze and obtain vehicle quantity data at a target intersection according to the image segmentation result, and to analyze and obtain first congestion degree information of the target intersection according to the vehicle quantity data; A second congestion analysis module is used to identify and obtain multiple headway distances between multiple vehicles in the target intersection according to the image segmentation result, and calculate and obtain second congestion degree information; The visualization identification module is used to identify and classify the congestion of the target intersection based on the first congestion information and the second congestion information, obtain the comprehensive congestion degree and the corresponding congestion identification information, and perform visualization identification on the target intersection.

9. A computer device, characterized in that: The computer device comprises a memory and a processor, wherein a computer program is stored in the memory, and when the computer program is executed by the processor, the steps of the method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores a computer program, and when the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 7 are implemented.

Images