A method and system for detecting the credibility of content published by a variable information board
By installing image acquisition devices on vehicles at highway entrances and using a detection server to perform variable message sign image matching and similarity calculation, the problems of insufficient real-time detection and automation in existing technologies are solved, achieving full coverage and low-cost variable message sign content detection.
Patent Information
- Application Number
- CN202210781012.7
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2022-07-04
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2042-07-04
AI Technical Summary
Existing technologies are insufficient in confirming the real-time nature and automation of variable message sign (VMS) content, and installing cameras involves a large amount of engineering work and is difficult to maintain.
By installing image acquisition devices on vehicles entering the highway, real-time images of variable message signs are captured, and a detection server is used to perform image matching and similarity calculations to determine the credibility of the published content, thus avoiding the need to set up a separate camera for each variable message sign.
It has achieved full coverage detection of variable message signs on highways, reducing engineering workload and maintenance costs, and improving the real-time performance and automation of detection.
Smart Images

Figure CN115035457B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of highway intelligent management, and in particular to a method and system for detecting the credibility of content published by a variable information board. Background Art
[0002] Variable information board is an important device for publishing traffic conditions and traffic guidance information. It is a traffic information display device with LED light-emitting devices as the basic display unit, and has graphic and text display functions. According to the traffic, weather and command information of the command and dispatch department, the operator can publish various notices and information to the variable information board, thereby effectively guiding the traffic flow, improving the transportation capacity of the road network, and providing high-quality services for drivers to drive safely and quickly.
[0003] After the current variable information board publishes content, the main methods to confirm that the content published by the information board is indeed issued by the background control system are:
[0004] (1) Manual confirmation through non-dedicated cameras near the variable information board;
[0005] (2) For variable information boards not covered by cameras, manual inspection is used to confirm;
[0006] (3) Obtain the published content of the variable information board through other means (such as feedback from drivers and inspectors, etc.) and manually confirm it.
[0007] The above methods have long confirmation cycles, poor real-time performance, and high costs.
[0008] The invention patent with application number 202210375371.2 discloses a method and system for detecting the credibility of the content published by a variable information board, the method comprising step 1: determining the content that the variable information board needs to publish, and controlling the variable information board to publish the content; step 2: obtaining the data of the content that the variable information board needs to publish, that is, the target image; step 3: obtaining the image data of the content being published by the variable information board in real time, that is, the real-time image; step 4: calculating the similarity between the real-time image and the target image, and judging the credibility of the content published by the variable information board; the system is used to execute the method. The invention can detect the credibility of the content published by the variable information board in real time, with good real-time performance; the whole detection process has a high degree of automation, which greatly reduces the labor cost; it can cover all information boards under the user's jurisdiction without omission; and it has an automatic alarm function. However, the system needs to set a camera for each variable information board that needs to be detected to collect real-time images, which is a large amount of engineering, and once the camera fails, it needs to be repaired on the spot, which increases the workload of the maintenance personnel. Summary of the invention
[0009] In order to solve at least one of the above technical problems, the present invention provides a method and system for detecting the credibility of the content published on variable information boards, which can realize the credibility detection of the content published on variable information boards on highways without setting up a separate camera for each variable information board that needs to be detected.
[0010] A first aspect of the present invention provides a method for detecting the credibility of content published by a variable information board, comprising:
[0011] Step 1: An image acquisition device is installed on a vehicle that enters the expressway through a manual toll lane, and is installed at a position on the vehicle where a front image can be acquired;
[0012] Step 2: The image acquisition device acquires real-time images of the content published by the variable information board and uploads them to the detection server;
[0013] Step 3: The detection server obtains the target image of the content to be published by the variable information board according to the play list uploaded by the play control device;
[0014] Step 4: The detection server matches the corresponding variable information board according to the position information contained in the real-time image, and then determines the target image at the position corresponding to the real-time image;
[0015] Step 5: The detection server matches the target image at the corresponding position according to the time information contained in the real-time image, and then determines the target image in the time period corresponding to the real-time image;
[0016] Step 6: The detection server calculates the similarity between the real-time image and the target image in the corresponding time period, and determines the credibility of the content published by the variable information board.
[0017] Preferably, the method further comprises the image acquisition device being recovered when the vehicle passes through a manual toll lane and leaves the expressway.
[0018] Preferably, in any of the above schemes, step 2 comprises:
[0019] Step 21: The image acquisition device acquires video images;
[0020] Step 22: The image acquisition device acquires the position coordinates thereof;
[0021] Step 23: The image acquisition device identifies whether there is a variable information board in the video image it acquires. If so, execute step 24; if not, continue to execute steps 21 and 22;
[0022] Step 24: The position coordinates are merged with the video image of the variable information board to obtain a real-time image of the content published by the variable information board containing the position information, and the real-time image is uploaded to the detection server.
[0023] Preferably, in any of the above schemes, step 4 comprises:
[0024] Step 41: Read the location information of the real-time image;
[0025] Step 42: Using the last moment position coordinates of the real-time image as a reference, find a set distance threshold range D = (d x ,d y ,d z ) within the variable information board, and then determine the variable information board at a position corresponding to the real-time image;
[0026] Step 43: Read the play list of the variable information board at the determined corresponding position, and then determine the target image at the position corresponding to the real-time image.
[0027] Preferably, in any of the above schemes, step 5 comprises:
[0028] Step 51: Read the time information of the real-time image to determine the start time and end time of the real-time image;
[0029] Step 52: Calculate the target image at the position corresponding to the real-time image determined in step 4 between the start time and the end time, and then determine the target image in the time period corresponding to the real-time image.
[0030] Preferably, in any of the above solutions, step 6 comprises:
[0031] Step 61: The detection server samples the real-time image at a fixed frequency starting from the start time to obtain a real-time image sampling image sequence I including l sampled images. a ={i a1 ,i a2 ,…,i al};
[0032] Step 62: The detection server samples the target image in the time period corresponding to the real-time image at the same starting time and at the fixed frequency to obtain a target image sampling sequence I including l sampled images. o ={i o1 ,i o2 ,…,i ol};
[0033] Step 63: The detection server calculates the similarity between the real-time image sampling graph and the target image sampling graph at the same time, and obtains a similarity value sequence S including l similarity values = {s1, s2, ..., s l};
[0034] Step 64: Take the average of the l similarity values to obtain the overall similarity
[0035] Step 65: Determine whether the overall similarity CS is greater than a set threshold value K. If so, it means that the content published by the variable information board is credible; otherwise, it means that the content published by the variable information board is not credible, and the detection server issues an alarm.
[0036] A second aspect of the present invention provides a system for detecting the credibility of content published by a variable information board, which is used to execute the method, and the system comprises:
[0037] Image acquisition equipment: used to collect real-time images of the content published by the variable information board and upload them to the detection server;
[0038] Playback control device: used to determine the content that needs to be published on the variable information board, control the variable information board to publish the content, and upload the playlist to the detection server;
[0039] Detection server: used to obtain the target image of the content that the variable information board needs to publish according to the play list uploaded by the playback control device, and judge the credibility of the content published by the variable information board according to the real-time image and the play list.
[0040] Preferably, the image acquisition device is disposed on a vehicle that enters the expressway through a manual toll lane, and is installed at a position on the vehicle where it can acquire images ahead.
[0041] Preferably, in any of the above solutions, the image acquisition device is recovered when the vehicle passes through the manual toll collection channel and leaves the expressway.
[0042] Preferably, in any of the above solutions, the image acquisition device includes a detachable mounting component for arranging it on a vehicle.
[0043] The method and system for detecting the credibility of the content published by the variable information board of the present invention have the following beneficial effects:
[0044] The image acquisition equipment carried by vehicles entering the expressway through the manual toll channel can be used to collect the content published by the variable information board on the expressway, thereby eliminating the need to set up a camera for each variable information board, thus reducing the amount of work and cost;
[0045] When the vehicle leaves the expressway, the image acquisition device is recovered and can be sent to a fixed maintenance center for centralized maintenance when the image acquisition device fails, thereby reducing the workload of maintenance personnel who need to repair the camera on site;
[0046] With the help of vehicles entering the expressway, the expressway network will be fully covered, and the variable information boards on the expressway can be fully covered with image acquisition, so that the credibility of the content released by all variable information boards on the expressway can be tested;
[0047] By taking advantage of the random distribution of vehicles on the highway network, it is possible to perform irregular random checks on the credibility of the content published by a single variable information board. BRIEF DESCRIPTION OF THE DRAWINGS
[0048] Figure 1 The figure is a flow chart of a preferred embodiment of a method for detecting the credibility of content published by a variable information board according to the present invention.
[0049] Figure 2 The figure is a schematic diagram of the structure of a preferred embodiment of a system for detecting the credibility of content published on a variable information board according to the present invention.
[0050] Figure 3 According to the method for detecting the credibility of the content published by the variable information board of the present invention, Figure 1 A schematic flow chart of step 2 of the illustrated embodiment.
[0051] Figure 4 According to the method for detecting the credibility of the content published by the variable information board of the present invention, Figure 1 A schematic flow chart of step 4 of the illustrated embodiment.
[0052] Figure 5 According to the method for detecting the credibility of the content published by the variable information board of the present invention, Figure 1 A schematic flow chart of step 5 of the illustrated embodiment.
[0053] Figure 6 According to the method for detecting the credibility of the content published by the variable information board of the present invention, Figure 1 A schematic flow chart of step 6 of the illustrated embodiment. DETAILED DESCRIPTION
[0054] In order to better understand the present invention, the present invention is described in detail below in conjunction with specific embodiments.
[0055] Example 1
[0056] like Figure 1 As shown, a method for detecting the credibility of content published by a variable information board includes:
[0057] Step 1: An image acquisition device is installed on a vehicle that enters the expressway through a manual toll lane, and is installed at a position on the vehicle where a front image can be acquired;
[0058] Step 2: The image acquisition device acquires real-time images of the content published by the variable information board and uploads them to the detection server;
[0059] Step 3: The detection server obtains the target image of the content to be published by the variable information board according to the play list uploaded by the play control device;
[0060] Step 4: The detection server matches the corresponding variable information board according to the position information contained in the real-time image, and then determines the target image at the position corresponding to the real-time image;
[0061] Step 5: The detection server matches the target image at the corresponding position according to the time information contained in the real-time image, and then determines the target image in the time period corresponding to the real-time image;
[0062] Step 6: The detection server calculates the similarity between the real-time image and the target image in the corresponding time period, and determines the credibility of the content published by the variable information board.
[0063] In this embodiment, preferably, the method further includes the image acquisition device being recovered when the vehicle passes through the manual toll channel and leaves the expressway. It should be noted that in this application, the image acquisition device is only set on the vehicle entering the expressway through the manual toll channel because the vehicle entering the expressway through the manual channel needs to stop at the toll station to exchange for CPC cards, and needs to stop at the toll station to pay manually when leaving the expressway, which provides an opportunity for installing and recovering the image acquisition device.
[0064] like Figure 2 As shown, a system for detecting the credibility of content published by a variable information board is used to execute the method, and the system includes:
[0065] Image acquisition device 10: used to acquire real-time images of the content published by the variable information board and upload them to the detection server 30;
[0066] Play control device 20: used to determine the content that needs to be published on the variable information board, control the variable information board to publish the content, and upload the play list to the detection server 30;
[0067] Detection server 30: used to obtain the target image of the content to be published by the variable information board according to the play list uploaded by the play control device 20, and judge the credibility of the content published by the variable information board according to the real-time image and the play list.
[0068] In this embodiment, it is preferred that the image acquisition device 10 includes a detachable mounting component for setting it on the vehicle. Specifically, the detachable mounting component can be a suction cup, by which the image acquisition device 10 is mounted on the center console or the front windshield of the vehicle; the detachable mounting component can also be a buckle, by which the image acquisition device 10 is mounted on the back of the rearview mirror of the vehicle. The detachable mounting component can also be designed as required, as long as the image acquisition device 10 can be easily installed and removed. The installation position of the image acquisition device 10 should ensure that the image acquisition device 10 can capture image information in front of the vehicle.
[0069] like Figure 3 As shown, the step 2 includes:
[0070] Step 21: The image acquisition device 10 acquires video images;
[0071] Step 22: The image acquisition device 10 acquires the coordinates of its location;
[0072] Step 23: The image acquisition device 10 identifies whether there is a variable information board in the video image it acquires. If so, execute step 24; if not, continue to execute steps 21 and 22;
[0073] Step 24 : The position coordinates are merged with the video image of the variable information board to obtain a real-time image of the content published by the variable information board containing the position information, and the real-time image is uploaded to the detection server 30 .
[0074] In this embodiment, it is preferred that the image acquisition device 10 uses a neural network to identify whether there is a variable information board in the collected video image. Since the credibility of the content published by the variable information board needs to be checked, it should be identified whether there is the front of the variable information board in the video image collected by the image acquisition device 10. Therefore, when training the neural network, the training set used is the front of the variable information board, thereby preventing the variable information board on the opposite lane from being misidentified. In step 24, only when the image acquisition device 10 identifies that there is a variable information board in the video image it has collected, the position coordinates are merged with the video image where the variable information board exists and uploaded, which can effectively save network resources and computing resources.
[0075] In this embodiment, preferably, in step 3, to save network resources, the playback control device 20 re-uploads the updated playlist to the detection server 30 only when the playlist changes. Steps 2 and 3 may be performed simultaneously, or step 3 may be performed first and then step 2.
[0076] like Figure 4 As shown, step 4 includes:
[0077] Step 41: The detection server 30 reads the location information of the real-time image;
[0078] Step 42: The detection server 30 uses the position coordinates of the real-time image at the last moment as a reference to find a set distance threshold D=(d x ,d y ,d z ) within the range, and then determine the variable information board at the position corresponding to the real-time image;
[0079] Step 43: The detection server 30 reads the play list of the variable information board at the determined corresponding position, and further determines the target image at the position corresponding to the real-time image.
[0080] It should be noted that in step 42, the position coordinates of the real-time image at the last moment are used as the reference because during the entire acquisition period of the real-time image, at the last moment, the image acquisition device 10 is closest to the position of the variable information board. The distance threshold D = (d x ,d y ,d z ) is a value set according to needs. Specifically, it can be set by comprehensively considering the road speed limit, whether there is an overpass, etc. It should be noted that if there is an overpass, there may be variable information boards at the same XY coordinates but different Z coordinates. In order to improve the matching accuracy, d z is a value range, which can ensure that only the variable information board above the road surface where the vehicle is traveling is positionally matched.
[0081] like Figure 5 As shown, step 5 includes:
[0082] Step 51: Read the time information of the real-time image to determine the start time and end time of the real-time image;
[0083] Step 52: Calculate the target image at the position corresponding to the real-time image determined in step 4 between the start time and the end time, and then determine the target image in the time period corresponding to the real-time image.
[0084] It should be noted that the real-time image contains time information when it is collected, that is, the time when the real-time image is collected. Then, in step 5, time matching can be directly performed based on the time information included in the real-time image to determine the target image of the time period corresponding to the real-time image, which can prevent delays in the process of uploading the real-time image, thereby causing time matching errors.
[0085] like Figure 6 As shown, step 6 includes:
[0086] Step 61: The detection server samples the real-time image at a fixed frequency starting from the start time to obtain a real-time image sampling image sequence I including l sampled images. a ={i a1 ,i a2 ,…,i al};
[0087] Step 62: The detection server samples the target image in the time period corresponding to the real-time image at the same starting time and at the fixed frequency to obtain a target image sampling sequence I including l sampled images. o ={i o1 ,i o2 ,…,i ol};
[0088] Step 63: The detection server calculates the similarity between the real-time image sampling graph and the target image sampling graph at the same time, and obtains a similarity value sequence S including l similarity values = {s1, s2, ..., s l};
[0089] Step 64: Take the average of the l similarity values to obtain the overall similarity
[0090] Step 65: Determine whether the overall similarity CS is greater than a set threshold value K. If so, it means that the content published by the variable information board is credible; otherwise, it means that the content published by the variable information board is not credible, and the detection server issues an alarm.
[0091] It should be noted that after step 5, the real-time image and the target image have the same start time and end time. In step 61, since the image acquisition device 10 is getting closer and closer to the variable information board during the process of acquiring the real-time image, the viewing angle of the acquired real-time image is constantly changing. Therefore, the real-time image can be corrected before sampling, or each frame of the image in the real-time image sampling diagram sequence obtained after sampling can be corrected. In steps 61 and 62, the real-time image and the target image are sampled at the same starting point and frequency, and the first frame of the real-time image sampling diagram i is obtained. a1 And the first frame target image sampling map i o1 At the same time, the second frame real-time image sampling diagram i a2 And the second frame target image sampling map i o2 The moment is the same, ... In this embodiment, preferably, in step 63, a convolutional neural network is used to calculate the similarity between the real-time image sampling graph and the target image sampling graph at the same moment. Since the convolutional neural network is a very mature and widely used method, it will not be introduced in detail in this application.
[0092] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the above embodiments describe the present invention in detail, those skilled in the art should understand that they may modify the technical solutions recorded in the above embodiments, or replace some or all of the technical features therein with equivalents, and these replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the present invention.
Claims
1. A method for detecting the credibility of content published by a variable information board, comprising step 3: a detection server obtains a target image of the content to be published by the variable information board according to a play list uploaded by a play control device, characterized in that: Also includes: Step 1: An image acquisition device is installed on a vehicle that enters the expressway through a manual toll lane, and is installed at a position on the vehicle where a front image can be acquired; Step 2: The image acquisition device acquires real-time images of the content published by the variable information board and uploads them to the detection server; Step 4: The detection server matches the corresponding variable information board according to the position information contained in the real-time image, and then determines the target image at the position corresponding to the real-time image; Step 5: The detection server matches the target image at the corresponding position according to the time information contained in the real-time image, and then determines the target image in the time period corresponding to the real-time image; Step 6: The detection server calculates the similarity between the real-time image and the target image in the corresponding time period, and determines the credibility of the content published by the variable information board.
2. The method for detecting the credibility of content published by a variable information board according to claim 1, characterized in that: It also includes the image acquisition device being recovered when the vehicle passes through the manual toll channel and leaves the expressway.
3. The method for detecting the credibility of content published by a variable information board according to claim 1, characterized in that: The step 2 comprises: Step 21: The image acquisition device acquires video images; Step 22: The image acquisition device acquires the position coordinates thereof; Step 23: The image acquisition device identifies whether there is a variable information board in the video image it acquires. If so, execute step 24; if not, continue to execute steps 21 and 22; Step 24: The position coordinates are merged with the video image of the variable information board to obtain a real-time image of the content published by the variable information board containing the position information, and the real-time image is uploaded to the detection server.
4. The method for detecting the credibility of content published by a variable information board as claimed in claim 3, characterized in that: Step 4 includes: Step 41: Read the location information of the real-time image; Step 42: Using the last moment position coordinates of the real-time image as a reference, find a set distance threshold range D = (d x ,d y ,d z ) within the variable information board, and then determine the variable information board at a position corresponding to the real-time image; Step 43: Read the play list of the variable information board at the determined corresponding position, and then determine the target image at the position corresponding to the real-time image.
5. The method for detecting the credibility of content published by a variable information board as claimed in claim 4, characterized in that: Step 5 includes: Step 51: Read the time information of the real-time image to determine the start time and end time of the real-time image; Step 52: Calculate the target image at the position corresponding to the real-time image determined in step 4 between the start time and the end time, and then determine the target image in the time period corresponding to the real-time image.
6. The method for detecting the credibility of content published by a variable information board as claimed in claim 5, characterized in that: The step 6 comprises: Step 61: The detection server samples the real-time image at a fixed frequency starting from the start time to obtain a real-time image sampling image sequence I including l sampled images. a ={i a1 ,i a2 ,…,i al }; Step 62: The detection server samples the target image in the time period corresponding to the real-time image at the same starting time and at the fixed frequency to obtain a target image sampling sequence I including l sampled images. o ={i o1 ,i o2 ,…,i ol }; Step 63: The detection server calculates the similarity between the real-time image sampling graph and the target image sampling graph at the same time, and obtains a similarity value sequence S including l similarity values = {s1, s2, ..., s l }; Step 64: Take the average of the l similarity values to obtain the overall similarity Step 65: Determine whether the overall similarity CS is greater than a set threshold value K. If so, it means that the content published by the variable information board is credible; otherwise, it means that the content published by the variable information board is not credible, and the detection server issues an alarm.
7. A system for detecting the credibility of content published by a variable information board, characterized in that: For executing the method according to any one of claims 1 to 6, the system comprises: Image acquisition equipment: used to collect real-time images of the content published by the variable information board and upload them to the detection server; Playback control device: used to determine the content that needs to be published on the variable information board, control the variable information board to publish the content, and upload the playlist to the detection server; Detection server: used to obtain the target image of the content that the variable information board needs to publish according to the playback list uploaded by the playback control device, and judge the credibility of the content published by the variable information board according to the real-time image and the playback list.
8. The system for detecting the credibility of content published by a variable information board as claimed in claim 7, characterized in that: The image acquisition device is arranged on a vehicle that enters the expressway through a manual toll collection channel, and is installed at a position on the vehicle where the image ahead can be acquired.
9. The system for detecting the credibility of content published by a variable information board as claimed in claim 8, characterized in that: The image acquisition device is recovered when the vehicle passes through the manual toll channel and leaves the expressway.
10. The system for detecting the credibility of content published by a variable information board as claimed in claim 8, characterized in that: The image acquisition device includes a detachable mounting component for arranging it on a vehicle.
