Traffic dispersion system based on image recognition technology

Abstract

The invention relates to the technical field of traffic dispersion, and discloses a traffic dispersion system based on an image recognition technology, and the system comprises an image collection module which is used for capturing image data in a traffic scene; the processing module comprises an image processing unit and an analysis unit, and the image processing unit is used for preprocessing the collected image data to obtain image feature indexes; the analysis unit is used for analyzing the characteristic indexes; the information management module comprises an information generation end and a traffic guidance screen, and the information generation end is used for generating a traffic dispersion information prompting strategy according to the analysis result; and the traffic guidance screen is arranged at the entrance end of the target road and is used for executing an information traffic dispersion information prompting strategy. According to the scheme, the visual field range of the traffic participants is expanded, so that when the traffic participants participate in traffic, the drivers can know the condition of the road ahead in time and take targeted response measures according to the prompt content of the traffic dispersion information, and the road passing efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of traffic diversion, and in particular to a traffic diversion system based on image recognition technology. Background Art

[0002] Traffic diversion refers to guiding, regulating and managing road traffic flow through a series of measures. Reasonable traffic diversion measures can alleviate traffic congestion, improve road traffic efficiency, reduce traffic delays and emissions, achieve orderly, safe, efficient and environmentally friendly operation of road traffic, and improve the quality of urban traffic operation.

[0003] However, in the face of increasingly serious traffic congestion and the limitations of existing technologies, it is necessary to continuously explore and innovate methods and means of traffic diversion to meet the needs of urban traffic development. The limitations of existing traffic diversion technologies are specifically reflected in the following aspects: it is difficult for traffic participants to predict whether the road ahead is congested, and when the road is congested, it is impossible for traffic participants to know the road ahead from their perspective. Therefore, it is difficult for traffic participants to respond in a timely manner according to the specific situation of the source of road congestion (such as: reasonable lane change, early deceleration, etc.). Summary of the invention

[0004] The purpose of the present invention is to provide a traffic diversion system based on image recognition technology to solve the above technical problems:

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

[0006] A traffic diversion system based on image recognition technology, comprising:

[0007] An image acquisition module, used to capture image data in traffic scenes;

[0008] The processing module includes an image processing unit and an analysis unit, wherein the image processing unit is used to pre-process the collected image data to obtain image feature indicators; and the analysis unit is used to analyze the feature indicators;

[0009] The information management module includes an information generation terminal and a traffic guidance screen. The information generation terminal is used to generate a traffic diversion information prompt strategy based on the analysis results; the traffic guidance screen is set at the entrance of the target road to execute the information traffic diversion information prompt strategy. The above technical solution provides a traffic diversion system based on image recognition technology. The image acquisition module obtains image data in the traffic scene, and then the processing module processes and analyzes the graphic data to generate a traffic diversion information prompt strategy, and the traffic guidance screen displays the traffic diversion prompt information to traffic participants. In summary, this technical solution expands the field of vision of traffic participants, so that traffic participants can promptly know the road conditions ahead, and make targeted response measures according to the content of the traffic diversion information prompt, thereby improving road traffic efficiency.

[0010] As a further technical solution, the image acquisition module includes:

[0011] Monitoring cameras, wherein there are multiple monitoring cameras, all of which are set on the target road section; a drone image acquisition unit, including a drone platform and a drone set on the target road section; the monitoring camera and the drone are both in communication with the processing module. The above technical solution provides the specific content of the image acquisition module, wherein there are multiple monitoring cameras, which can realize interval speed measurement on the one hand, and improve the fault tolerance of image acquisition on the other hand. The drone platform provides parking, charging, inspection and other services for the drone.

[0012] As a further technical solution, the process of preprocessing the collected image data includes:

[0013] Perform image enhancement processing with unified standard parameters on image data and remove noise in the image through filters;

[0014] After scaling the image to a preset size, the trained convolutional neural network model is imported, and the output result obtains characteristic indicators including: the number of vehicles in each lane in the image, the number of vehicles changing lanes, the direction of vehicle lane change, and the speed of the vehicle. The above technical solution provides a process for preprocessing the collected image data. Among them, the purpose of image enhancement processing is to enhance certain features of the image so that the content of the image can be observed more clearly. It can be achieved by enhancing edges, improving contrast, correcting colors, etc. The main purpose of image denoising is to remove unnecessary random signals from images interfered by noise to restore the true content of the image. After scaling the image to a preset size, the trained convolutional neural network model is imported, and the output result obtains characteristic indicators including: the number of vehicles in each lane in the image, the number of vehicles changing lanes, the direction of vehicle lane change, and the speed of the vehicle.

[0015] As a further technical solution, the process of analyzing the characteristic indicators includes:

[0016] S1. Comprehensively evaluate and analyze the target road based on the characteristic indicators in the image obtained by the surveillance camera;

[0017] S2. Determine whether the target road is congested based on the comprehensive evaluation and analysis results, generate corresponding road prompt information, and display it on the traffic guidance screen;

[0018] S3. If the target road is judged to be in a congested state, dynamic monitoring and analysis are performed on the target road, and the dynamic monitoring and analysis result indicates whether to dispatch a drone to collect local images of the target road;

[0019] S4. Import the image data collected by the drone into the trained recognition model, output the information indicating the source of road congestion, and display it on the traffic guidance screen.

[0020] The above technical solution provides a process for analyzing characteristic indicators. It clarifies the order in which surveillance cameras and drones collect image data, and selectively triggers the command to dispatch drones based on the analysis results of the images captured by the surveillance cameras. It realizes the overall evaluation and local key analysis of the target road.

[0021] As a further technical solution, the process of comprehensive evaluation and analysis of the target road includes:

[0022]

[0023]

[0024] P = k1y1 + k2y2 (3);

[0025] In formulas (1) to (3), y1 and y2 are characteristic parameters; c i is the number of vehicles passing through the i-th lane captured by the surveillance camera within the preset unit time; n is the number of lanes in the same direction of the target section, and w i is the preset lane weight coefficient; v j is the speed of the jth vehicle passing through in the preset unit time; m is the total number of vehicles passing through in the preset unit time; v st is the preset standard vehicle speed; k1 and k2 are preset characteristic weight coefficients, and k2>k1;

[0026] The congestion risk coefficient P of the target road is calculated by formulas (1) to (3);

[0027] If the congestion risk factor P exceeds the preset risk threshold P max , the target road is judged to be in a congested state, and a warning signal of road congestion is issued through the traffic guidance screen; otherwise, no warning signal is issued.

[0028] As a further technical solution, the process of dynamic monitoring and analysis of the target road includes:

[0029] When P>P max After the analysis results are obtained, the current congestion risk coefficient is recorded as P1;

[0030] After that, the congestion risk coefficients of the preset unit time periods are obtained continuously for N times and marked as P2, P3…P N-1 .

[0031] As a further technical solution, the process of dynamic monitoring and analysis of the target road also includes:

[0032] Get the congestion risk coefficient values ​​P1~P N and the right endpoints t1 to t2 corresponding to the preset unit time period N ;

[0033] Draw P1~P N The change image is obtained, and the risk factor change curve is obtained through smoothing processing. The slope of the risk factor change curve is monitored. When the slope exceeds the preset upper limit, a command to dispatch a drone is issued to the drone platform.

[0034] As a further technical solution, the road prompt information includes target road congestion prompt information; the congestion source information includes: road construction location, traffic accident location, and traffic signal light failure location.

[0035] Beneficial effects of the present invention:

[0036] The present invention acquires image data in the traffic scene through an image acquisition module, and then the processing module processes and analyzes the graphic data, and finally generates a traffic diversion information prompt strategy, and the traffic guidance screen displays the traffic diversion prompt information to traffic participants. The scheme of the present invention expands the field of vision of traffic participants, so when participating in traffic, the driver can timely know the road conditions ahead, and take targeted response measures according to the content of the traffic diversion information prompt, thereby improving the road traffic efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0037] The present invention will be further described below in conjunction with the accompanying drawings.

[0038] Figure 1 A block diagram showing the contents of the traffic diversion system based on image recognition technology in the present invention;

[0039] Figure 2 The figure is a flow chart of the system in the present invention analyzing characteristic indicators. DETAILED DESCRIPTION

[0040] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not 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.

[0041] See also Figure 1 As shown, a traffic diversion system based on image recognition technology includes:

[0042] An image acquisition module, used to capture image data in traffic scenes;

[0043] The processing module includes an image processing unit and an analysis unit. The image processing unit is used to pre-process the collected image data to obtain image feature indicators; the analysis unit is used to analyze the feature indicators; the information management module includes an information generation end and a traffic guidance screen. The information generation end is used to generate a traffic diversion information prompt strategy according to the analysis results; the traffic guidance screen is set at the entrance of the target road and is used to execute the information traffic diversion information prompt strategy. The traffic diversion information prompt content may specifically include: road congestion ahead, road congestion reasons, passable lanes, and early lane change suggestions. Through the above technical solution, this embodiment provides a traffic diversion system based on image recognition technology. Specifically, the image data in the traffic scene is obtained through the image acquisition module, and then the processing module processes and analyzes the graphic data, and finally generates a traffic diversion information prompt strategy, and the traffic guidance screen displays the traffic diversion prompt information to the traffic participants. In summary, this technical solution expands the field of vision of traffic participants, so that traffic participants can timely know the road conditions ahead and make targeted response measures according to the traffic diversion information prompt content.

[0044] The image acquisition module includes: a monitoring camera, wherein a plurality of monitoring cameras are provided and all are arranged on the target road section; a drone image acquisition unit, including a drone platform and a drone arranged on the target road section; and the monitoring camera and the drone both establish a communication connection relationship with the processing module.

[0045] Through the above technical solution, this embodiment provides the specific content of the image acquisition module, wherein multiple surveillance cameras are provided, which can realize interval speed measurement on the one hand, and improve the fault tolerance of image acquisition on the other hand. The drone platform provides parking, charging and inspection services for the drone.

[0046] The process of preprocessing the acquired image data includes:

[0047] Perform image enhancement processing with unified standard parameters on image data and remove noise in the image through filters;

[0048] After scaling the image to a preset size, the trained convolutional neural network model is imported, and the output results obtain characteristic indicators including: the number of vehicles in each lane in the image, the number of vehicles changing lanes, the direction of vehicle lane changes, and the vehicle speed.

[0049] Through the above technical solution, this embodiment provides a process for preprocessing the collected image data. Specifically, the image data is first subjected to image enhancement processing with unified standard parameters, and the noise in the image is removed by a filter. The purpose of image enhancement processing is to enhance certain features of the image so that the content of the image can be observed more clearly. It can be achieved specifically by enhancing edges, improving contrast, correcting colors, etc. The main purpose of image denoising is to remove unnecessary random signals from images interfered by noise to restore the true content of the image. After scaling the image to a preset size, the trained convolutional neural network model is imported, and the output result obtains characteristic indicators including: the number of vehicles in each lane in the image, the number of vehicles changing lanes, the direction of vehicle lane change, and the speed of the vehicle. Among them, the training of the convolutional neural network can be completed through experiments, which is a prior art and will not be described in detail here.

[0050] See also Figure 2 As shown in Figure 1, the process of analyzing the characteristic indicators includes:

[0051] S1. Comprehensively evaluate and analyze the target road based on the characteristic indicators in the image obtained by the surveillance camera;

[0052] S2. Determine whether the target road is congested based on the comprehensive evaluation and analysis results, generate corresponding road prompt information, and display it on the traffic guidance screen;

[0053] S3. If the target road is judged to be in a congested state, dynamic monitoring and analysis are performed on the target road, and the dynamic monitoring and analysis result indicates whether to dispatch a drone to collect local images of the target road;

[0054] S4. Import the image data collected by the drone into the trained recognition model, output the information indicating the source of road congestion, and display it on the traffic guidance screen.

[0055] Through the above technical solution, this embodiment provides a process for analyzing characteristic indicators. The order in which the surveillance camera and the drone collect image data is clarified, and the command to dispatch the drone is selectively triggered according to the analysis results of the images captured by the surveillance camera. The process of overall evaluation and local key analysis of the target road is realized. Specifically, according to the characteristic indicators in the image acquired by the surveillance camera, including the number of vehicles on the road, the number of vehicles changing lanes, the direction of vehicle lane change, and the speed of the vehicle. Then, a comprehensive evaluation and analysis is performed on the target road, and whether the target road is congested is determined according to the results of the comprehensive evaluation and analysis, and road prompt information is generated accordingly, and displayed on the traffic guidance screen. If the target road is judged to be in a congested state, the target road is dynamically monitored and analyzed, and the results of the dynamic monitoring and analysis indicate whether to dispatch a drone to collect local images of the target road. The image data collected by the drone is imported into the trained recognition model, and the output result indicates the source information of the road congestion, which is displayed on the traffic guidance screen.

[0056] The process of comprehensive evaluation and analysis of the target road includes:

[0057]

[0058] P = k1y1 + k2y2 (3);

[0059] In formulas (1) to (3), y1 and y2 are characteristic parameters; c i is the number of vehicles passing through the i-th lane captured by the surveillance camera within the preset unit time; n is the number of lanes in the same direction of the target section, and w i is the preset lane weight coefficient. Since each lane has a different proportion of vehicles under normal traffic conditions, the weight coefficient w of each lane is obtained based on the historical data during normal traffic. i ;v j is the speed of the jth vehicle passing through in the preset unit time; m is the total number of vehicles passing through in the preset unit time; v st is the preset standard vehicle speed; k1 and k2 are preset characteristic weight coefficients, which represent the uneven distribution of vehicles in each lane in the same direction and the different degrees of road congestion judgment due to speed reduction. The specific results can be obtained based on historical data fitting experiments and will not be described in detail here.

[0060] The congestion risk coefficient P of the target road is calculated by formulas (1) to (3);

[0061] If the congestion risk factor P exceeds the preset risk threshold P max , the target road is judged to be in a congested state, and a warning signal of road congestion is issued through the traffic guidance screen; otherwise, no warning signal is issued.

[0062] Through the above technical solution, this embodiment provides a process for comprehensive evaluation and analysis of the target road. Specifically, the congestion risk coefficient P of the target road is calculated by formulas (1) to (3), and then the congestion risk coefficient P is compared with the preset risk threshold P. max Compare, if P>P max , the target road is judged to be in a congested state, and a traffic congestion warning signal is issued through the traffic guidance screen; if P≤P max , no prompt signal is issued.

[0063] The process of dynamic monitoring and analysis of the target road includes:

[0064] When P>P max After the analysis results are obtained, the current congestion risk coefficient is recorded as P1;

[0065] After that, the congestion risk coefficients of the preset unit time periods are obtained continuously for N times and marked as P2, P3…P N-1 .

[0066] Through the above technical solution, this embodiment provides a process for dynamically monitoring and analyzing the target road. Specifically, when P>P is obtained through comprehensive evaluation analysis for the first time, max After the analysis result is obtained, the current congestion risk coefficient is recorded as P1; the congestion risk coefficients of the preset unit time periods are obtained N times continuously and marked as P2, P3…P N-1 By comparing P1 to P N , it can reflect the change of congestion risk coefficient.

[0067] The process of dynamic monitoring and analysis of the target road also includes:

[0068] Get the congestion risk coefficient values ​​P1~P N and the right endpoints t1 to t2 corresponding to the preset unit time period N ;

[0069] Draw P1~P N The change image is obtained, and the risk factor change curve is obtained through smoothing processing. The slope of the risk factor change curve is monitored. When the slope exceeds the preset upper limit, a command to dispatch a drone is issued to the drone platform.

[0070] Through the above technical solution, this embodiment provides a process for dynamically monitoring and analyzing the target road. Specifically, the congestion risk coefficient values ​​P1 to P N and the right endpoints t1 to t2 corresponding to the preset unit time period N , then draw P1~P NThe image of the change of the risk factor is obtained by smoothing, and the slope of the risk factor change curve is monitored. When the slope exceeds the preset upper limit, a command to dispatch a drone is issued to the drone platform. The preset upper limit can be set to a negative value, that is, when the risk factor does not decrease within a period of time, a drone is dispatched to conduct a local road condition survey.

[0071] The road prompt information includes target road congestion prompt information; the congestion source information includes: road construction location, traffic accident location, and traffic signal light failure location.

[0072] Through the above technical solution, this embodiment provides the specific content of road prompt information and congestion source information. The drone can quickly reach the accident or fault scene on the target road, and provide detailed information about the accident or fault, such as location, type, severity, etc., by taking high-definition photos or videos. Traffic participants can make reasonable responses based on this information, such as changing lanes in advance, slowing down, planning new routes, etc.

[0073] The present invention may be a system, a method and / or a computer program product. The computer program product may include a computer-readable storage medium carrying computer-readable program instructions for causing a processor to implement various aspects of the present invention.

[0074] A computer-readable storage medium may be a tangible device that can hold and store instructions used by an instruction execution device. A computer-readable storage medium may be, for example, but not limited to, an electrical storage device, a magnetic storage device, an optical storage device, an electromagnetic storage device, a semiconductor storage device, or any suitable combination of the foregoing. More specific examples of computer-readable storage media (a non-exhaustive list) include: a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM or flash memory), a static random access memory (SRAM), a portable compact disk read-only memory (CD-ROM), a digital versatile disk (DVD), a memory stick, a floppy disk, a mechanical encoding device, such as a punch card or a raised structure in a groove on which instructions are stored, and any suitable combination of the foregoing. As used herein, a computer-readable storage medium is not to be interpreted as a transient signal per se, such as a radio wave or other freely propagating electromagnetic wave, an electromagnetic wave propagating through a waveguide or other transmission medium (e.g., a light pulse through a fiber optic cable), or an electrical signal transmitted through a wire.

[0075] The computer-readable program instructions described herein can be downloaded from a computer-readable storage medium to each computing / processing device, or downloaded to an external computer or external storage device via a network, such as the Internet, a local area network, a wide area network, and / or a wireless network. The network can include copper transmission cables, optical fiber transmissions, wireless transmissions, routers, firewalls, switches, gateway computers, and / or edge servers. The network adapter card or network interface in each computing / processing device receives the computer-readable program instructions from the network and forwards the computer-readable program instructions for storage in the computer-readable storage medium in each computing / processing device.

[0076] Although the present invention is disclosed as above, the protection scope of the present invention is not limited thereto. Those skilled in the art may make various changes and modifications without departing from the spirit and scope of the present invention, and these changes and modifications will fall within the protection scope of the present invention.

Claims

1. A traffic diversion system based on image recognition technology, characterized in that: The system comprises: An image acquisition module, used to capture image data in traffic scenes; The processing module includes an image processing unit and an analysis unit, wherein the image processing unit is used to pre-process the collected image data to obtain image feature indicators; and the analysis unit is used to analyze the feature indicators; The information management module includes an information generation terminal and a traffic guidance screen. The information generation terminal is used to generate a traffic diversion information prompt strategy according to the analysis results; the traffic guidance screen is set at the entrance of the target road and is used to execute the traffic diversion information prompt strategy.

2. A traffic diversion system based on image recognition technology according to claim 1, characterized in that: The image acquisition module comprises: A monitoring camera, wherein a plurality of monitoring cameras are provided, and all of the monitoring cameras are provided on the target road section; A drone image acquisition unit, including a drone platform and a drone set up on a target road section; The monitoring camera and the drone both establish a communication connection relationship with the processing module.

3. A traffic diversion system based on image recognition technology according to claim 2, characterized in that: The process of preprocessing the acquired image data includes: Perform image enhancement processing with unified standard parameters on image data and remove noise in the image through filters; After scaling the image to a preset size, the trained convolutional neural network model is imported, and the output results obtain characteristic indicators including: the number of vehicles in each lane in the image, the number of vehicles changing lanes, the direction of vehicle lane changes, and the vehicle speed.

4. The traffic diversion system based on image recognition technology according to claim 3 is characterized in that: The process of analyzing characteristic indicators includes: S1. Comprehensively evaluate and analyze the target road based on the characteristic indicators in the image obtained by the surveillance camera; S2. Determine whether the target road is congested based on the comprehensive evaluation and analysis results, generate corresponding road prompt information, and display it on the traffic guidance screen; S3. If the target road is judged to be in a congested state, dynamic monitoring and analysis are performed on the target road, and the dynamic monitoring and analysis result indicates whether to dispatch a drone to collect local images of the target road; S4. Import the image data collected by the drone into the trained recognition model, output the information indicating the source of road congestion, and display it on the traffic guidance screen.

5. The traffic diversion system based on image recognition technology according to claim 4 is characterized in that: The process of comprehensive evaluation and analysis of the target road includes: P = k1y1 + k2y2 (3); In formulas (1) to (3), y1 and y2 are characteristic parameters; c i is the number of vehicles passing through the i-th lane captured by the surveillance camera within the preset unit time; n is the number of lanes in the same direction of the target section, and w i is the preset lane weight coefficient; v j is the speed of the jth vehicle passing through in the preset unit time; m is the total number of vehicles passing through in the preset unit time; v st is the preset standard vehicle speed; k1 and k2 are the preset characteristic weight coefficients; The congestion risk coefficient P of the target road is calculated by formulas (1) to (3); If the congestion risk factor P exceeds the preset risk threshold P max , the target road is judged to be in a congested state, and a warning signal of road congestion is issued through the traffic guidance screen; otherwise, no warning signal is issued.

6. The traffic diversion system based on image recognition technology according to claim 5, characterized in that: The process of dynamic monitoring and analysis of the target road includes: When P>P max After the analysis results are obtained, the current congestion risk coefficient is recorded as P1; After that, the congestion risk coefficients of the preset unit time periods are obtained continuously for N times and marked as P2, P3…P N-1 .

7. The traffic diversion system based on image recognition technology according to claim 6 is characterized in that: The process of dynamic monitoring and analysis of the target road also includes: Get the congestion risk coefficient values ​​P1~P N and the right endpoints t1 to t2 corresponding to the preset unit time period N ; Draw P1~P N The change image is obtained, and the risk factor change curve is obtained through smoothing processing. The slope of the risk factor change curve is monitored. When the slope exceeds the preset upper limit, a command to dispatch a drone is issued to the drone platform.

8. The traffic diversion system based on image recognition technology according to claim 4 is characterized in that: The road prompt information includes target road congestion prompt information; the congestion source information includes: road construction location, traffic accident location, and traffic signal light failure location.

Images