A vehicle violation identification method and system based on a department edge machine

Abstract

This invention discloses a vehicle violation identification method and system based on a vehicle identification camera. The method involves acquiring the first vehicle's first speed and a first image of the area surrounding the first vehicle using a pre-set vehicle identification camera on the first vehicle. Then, based on the first speed, the method obtains the second speed of a second vehicle located around the first vehicle, thereby determining whether the second vehicle has committed a violation. A second image of the violation is saved, and finally, the image is analyzed a second time to obtain the violation information of the second vehicle. This improves the accuracy of vehicle violation identification and provides more comprehensive vehicle violation information.

Description

Technical Field

[0001] This invention relates to the field of smart cities, and in particular to a method and system for identifying vehicle violations based on a mobile terminal. Background Technology

[0002] With the rapid development of the information age and the gradual improvement of people's living standards, motor vehicles have become a more common means of transportation. However, many traffic safety problems have also become increasingly prominent, among which illegal parking and speeding on highways are significant safety hazards. Currently, highway vehicle monitoring still requires a large investment of manpower and resources, and the high volume of traffic during holidays is also a cause for concern, especially as unfavorable weather conditions such as heavy fog increase the probability of accidents on highways. Therefore, how to efficiently and accurately detect vehicle movement on highways has become an urgent problem to be solved. The current mainstream method is to use digital image processing technology for detection, but this is limited by the area captured by cameras on highways.

[0003] Currently, there are three main methods for capturing speeding and illegal parking. The first is fixed-location capture, which analyzes surveillance video images to determine whether a vehicle is speeding or illegally parked, and captures the violation if either occurs. The second is interval time difference speed measurement, where the command center database contains reasonable travel times for a given area. If the travel time through the area is less than a preset value, the vehicle is judged to be speeding and captured. The third is high-speed mobile speed measurement, generally used on mobile speed measurement vehicles. This type of speed detector creates a warning zone of a certain size in front of the radar, and captures the violation when a vehicle's speed exceeds the limit.

[0004] However, existing technology has significant limitations in using fixed camera locations to capture speeding or illegally parked vehicles. Even with mobile speed measurement, fixed locations are still required for capture, and the coverage area of ​​cameras on highways cannot be fully covered. Furthermore, current in-vehicle navigation systems prompt drivers to slow down in areas where cameras are being monitored, thus making it difficult to accurately determine whether a vehicle is illegally parked or speeding. Secondly, due to signal issues in tunnels, speeding or illegal parking cannot be effectively captured. Summary of the Invention

[0005] This invention discloses a vehicle violation identification method and system based on a mobile terminal, which improves the accuracy of vehicle violation identification and obtains more comprehensive vehicle violation information.

[0006] To achieve the above objectives, the present invention provides a vehicle violation identification method based on a mobile terminal, comprising:

[0007] The first vehicle speed and several first images of the road where the first vehicle is located are obtained by a pre-installed device in the first vehicle.

[0008] Based on the first image, obtain information about the second vehicle;

[0009] Based on the first vehicle speed and the information of the second vehicle, the second vehicle speed is obtained through a preset speed calculation formula and compared with the first vehicle speed to determine whether the second vehicle has committed a violation.

[0010] If the second vehicle violates the rules, multiple video trajectories corresponding to several second images containing the second vehicle are saved, and the second vehicle in the several second images is identified as the same vehicle in turn by a preset vehicle re-identification method.

[0011] If the second vehicle in the second image is the same vehicle, then the merged video is obtained by comparing the time interval and distance between the multiple video trajectories with a preset trajectory threshold.

[0012] The merged video is identified using a preset violation information identification method to obtain the violation information of the second vehicle.

[0013] This invention discloses a vehicle violation identification method based on a vehicle edge camera. First, a vehicle edge camera installed on a first vehicle acquires the vehicle's speed and the specific road conditions. Due to the stability of the edge camera, it can accurately and promptly obtain the vehicle's speed and road conditions, improving the stability of violation identification. Next, based on the first image, various information about a second vehicle is obtained for subsequent identification. After obtaining the speed of the first vehicle and the lane and relative position of the second vehicle, a preset speed calculation formula is used to process the lane and relative position, allowing the speed of the second vehicle to be obtained according to different lanes and distances, improving the accuracy of the second vehicle speed calculation and further enhancing the accuracy of violation identification. The second vehicle's second speed and type are obtained and compared with the speed limit corresponding to the first speed and the type to determine whether the second vehicle has committed illegal parking or speeding. The vehicle is classified to avoid errors in violation identification and improve the accuracy of identification. At the same time, the stability of the first speed measurement ensures the accuracy of the judgment, thereby improving the accuracy of identification. When it is determined that the second vehicle in the first image has committed a violation, the images of the violation are merged. Then, the merged video is identified to determine whether the second vehicle in the merged violation images is the same vehicle. Then, the second images of the same vehicle are merged to obtain a merged video. The violation information of the second vehicle that committed the violation is obtained from the merged video, thus obtaining the complete violation information of the second vehicle.

[0014] As a preferred example, the acquisition of a first vehicle speed and several first images of the road where the first vehicle is located by a pre-installed peripheral device in the first vehicle specifically includes:

[0015] The first vehicle speed is obtained by detecting the satellite inertial navigation and inertial measurement units preset in the device.

[0016] The device uses an ADAS camera pre-installed in the device to capture several first images of the road where the first vehicle is located.

[0017] This invention, through the satellite inertial navigation and inertial measurement units preset in the device, ensures that the speed of the first vehicle is obtained without being affected by geographical location, thus ensuring the accuracy of the first vehicle speed measurement. At the same time, based on the ADAS function of the device, it ensures that the first vehicle can accurately and unaffectedly capture the specific information of the road where the first vehicle is currently located. The stability of the ADAS ensures the stability of the capture. Based on the accuracy of the first vehicle speed, the accuracy of the second vehicle identification is improved.

[0018] As a preferred example, obtaining information about the second vehicle based on the first image specifically includes:

[0019] The ADAS camera analyzes the first image to obtain information about the second vehicle. The information about the second vehicle includes the lane in which the second vehicle is located, the type of the second vehicle, the license plate, and the relative position of the second vehicle to the first vehicle. The lane includes the first lane in which the first vehicle is located and the second lane adjacent to the first vehicle.

[0020] The present invention obtains a first image based on the ADAS and obtains specific information about the second vehicle, including the lane in which the second vehicle is located and the relative position of the second vehicle to the first vehicle, so as to facilitate subsequent determination of whether the second vehicle is speeding or illegally parked based on the information.

[0021] As a preferred example, obtaining the relative speed of the second vehicle using a preset relative speed calculation method specifically includes:

[0022] Based on the lane in which the second vehicle is located, determine whether the second vehicle is in the first lane or the second lane;

[0023] If the second vehicle is in the first lane, the first distance between the first vehicle and the second vehicle is obtained through the ADAS camera, and the first distance is processed by a preset first speed formula to obtain the second speed of the second vehicle.

[0024] If the second vehicle is in the second lane, the second distance between the first vehicle and the second vehicle and the lane line and the third distance between the first vehicle and the second vehicle are obtained through the ADAS camera. The second distance and the third distance are processed by a preset second speed formula to obtain the second speed of the second vehicle.

[0025] This invention uses different speed calculation methods to obtain the speed of the second vehicle based on the different lanes in which the second vehicle is located. According to the lane classification, the distance between the second vehicle and the first vehicle when the second vehicle is in different lanes is obtained, which improves the accuracy of the distance calculation between the first and second vehicles and the accuracy of the speed calculation of the second vehicle. This provides more accurate data for subsequent judgment of whether the vehicle has illegally parked or exceeded the speed limit, and further improves the accuracy of identification.

[0026] As a preferred example, the comparison with the first vehicle speed to determine whether the second vehicle has committed a violation specifically includes:

[0027] Based on the second vehicle's second speed and type, a comparison is made with the speed limit corresponding to the type of the second vehicle to determine whether the second vehicle is speeding;

[0028] If the second vehicle's second speed is greater than or equal to the speed limit, then the second vehicle is determined to be speeding.

[0029] Based on the first vehicle speed and the second vehicle speed, the relative speed of the second vehicle is obtained, and by comparing the relative speed with the first vehicle speed, it is determined whether the second vehicle has illegally parked.

[0030] If the relative speed is equal to the first vehicle speed, then it is determined that the second vehicle has illegally parked.

[0031] This invention calculates the relative speed of the second vehicle relative to the first vehicle based on the obtained second vehicle speed and the first vehicle speed. It then determines whether the second vehicle is illegally parked based on this relative speed. By leveraging the accuracy of the first vehicle speed, the accuracy of illegal parking identification is improved. Furthermore, when determining whether the second vehicle is speeding, the invention first obtains the type of the second vehicle and the corresponding speed limit for that type. Then, it compares the second vehicle's second speed with the speed limit to identify whether the second vehicle is speeding. By using the type of the second vehicle to determine whether it is speeding, the accuracy of speeding identification is improved.

[0032] As a preferred example, the step of sequentially identifying whether the second vehicles in the plurality of second images are the same vehicle using a preset vehicle re-identification method specifically includes:

[0033] By performing fine segmentation and secondary analysis on the second image, it is determined whether the second vehicle in the second image is driving illegally;

[0034] If the second vehicle is found to be driving illegally, the multiple video tracks corresponding to the second image are merged. At the same time, based on the license plate and features of the second vehicle in the second image, the second vehicle is identified using a preset vehicle re-identification technology to determine whether the second vehicle is the same vehicle.

[0035] This invention performs further fine segmentation and secondary analysis on the second image containing the second vehicle where the traffic violation occurred, to determine whether the traffic violation actually occurred in the second image, thereby improving the accuracy of recognition. Then, if the traffic violation did occur in the second image, it identifies whether the second vehicle in the second image is the same vehicle, thereby obtaining comprehensive information on the traffic violation of the second vehicle that committed the violation.

[0036] As a preferred example, the process of comparing the time intervals and distances between the multiple video trajectories with a preset trajectory threshold to obtain merged video recordings specifically includes:

[0037] If it is determined that the second vehicle is the same vehicle, then the time interval and distance between the several video tracks corresponding to the second image are compared with the preset time threshold and distance threshold to determine whether the second image should be merged.

[0038] If the time interval between the plurality of video tracks does not exceed the time threshold and the distance between them does not exceed the distance threshold, then the plurality of second images are merged to obtain a merged video.

[0039] If the time interval between the plurality of video tracks exceeds the time threshold or the distance between them exceeds the distance threshold, then the plurality of second images will not be merged.

[0040] This invention determines whether several video recordings constitute a traffic violation occurring at the same time and along the same route by setting time and distance thresholds. Two tracks that are too far apart do not belong to the same trip, and two tracks that are too far apart in time do not belong to the same trip. More accurate information about the traffic violation of the second vehicle can be obtained through the time and distance thresholds.

[0041] As a preferred example, the step of identifying the merged video using a preset violation identification method to obtain the violation information of the merged video specifically includes:

[0042] The radius of the video trajectory corresponding to the merged video is obtained by wrapping the merged video.

[0043] Based on the radius of the video trajectory, it is compared with a preset radius threshold to determine whether the merged video is a speeding video or a parking violation video, and at the same time, the violation information of the second vehicle is obtained based on the merged video.

[0044] This invention divides the merged video recordings into illegal parking and speeding recordings by setting a radius threshold, which facilitates the classification of violations by the vehicle on different road sections and at different times, and obtains more comprehensive violation information of the second vehicle.

[0045] On the other hand, the present invention provides a vehicle violation identification system based on a side-mounted machine, including an acquisition module, an information module, a judgment module, an identification module, a merging module and an output module;

[0046] The acquisition module is used to acquire the first vehicle speed and several first images of the road where the first vehicle is located through a pre-installed peripheral device in the first vehicle.

[0047] The information module is used to obtain information about the second vehicle based on the first image;

[0048] The judgment module is used to obtain the second speed of the second vehicle based on the first vehicle speed and the information of the second vehicle through a preset speed calculation formula, and compare it with the first vehicle speed to determine whether the second vehicle has committed a violation.

[0049] The identification module is used to save multiple video trajectories corresponding to several second images where the second vehicle is located if the second vehicle violates the rules, and to identify whether the second vehicle in the several second images is the same vehicle in turn through a preset vehicle re-identification method.

[0050] The merging module is used to obtain merged video recordings by comparing the time interval and distance between the multiple video trajectories with a preset trajectory threshold if the second vehicle in the second image is the same vehicle.

[0051] The output module is used to identify the merged video recording using a preset violation information identification method to obtain the violation information of the second vehicle.

[0052] This invention discloses a vehicle violation recognition system based on a vehicle edge camera. The system acquires the first vehicle's speed and the specific road conditions of the first vehicle using an edge camera installed on a first vehicle. Leveraging the stability of the edge camera, it accurately and promptly obtains the vehicle's speed and road conditions, improving the stability of violation recognition. Then, based on the first image, various information about a second vehicle is obtained for subsequent recognition. After obtaining the speed of the first vehicle and the lane and relative position of the second vehicle, a preset speed calculation formula is used to process the lane and relative position, allowing the speed of the second vehicle to be calculated based on different lanes and distances, improving the accuracy of the second vehicle speed calculation and further enhancing the accuracy of violation recognition. The second vehicle's second speed and type are compared with the speed limit corresponding to the first speed and the type to determine whether the second vehicle has committed illegal parking or speeding. The vehicle is classified to avoid errors in violation identification and improve the accuracy of identification. At the same time, the stability of the first speed measurement ensures the accuracy of the judgment, thereby improving the accuracy of identification. When it is determined that the second vehicle in the first image has committed a violation, the images of the violation are merged. Then, the merged video is identified to determine whether the second vehicle in the merged violation images is the same vehicle. Then, the second images of the same vehicle are merged to obtain a merged video. The violation information of the second vehicle that committed the violation is obtained from the merged video, thus obtaining the complete violation information of the second vehicle.

[0053] As a preferred example, the acquisition module includes a vehicle speed unit and an image unit;

[0054] The vehicle speed unit is used to detect and obtain the first vehicle speed of the first vehicle through the satellite inertial navigation and inertial measurement unit preset in the device.

[0055] The image unit is used to capture several first images of the road where the first vehicle is located through a pre-set ADAS camera in the device.

[0056] This invention, through the satellite inertial navigation and inertial measurement units preset in the device, ensures that the speed of the first vehicle is obtained without being affected by geographical location, thus ensuring the accuracy of the first vehicle speed measurement. At the same time, based on the ADAS function of the device, it ensures that the first vehicle can accurately and unaffectedly capture the specific information of the road where the first vehicle is currently located. The stability of the ADAS ensures the stability of the capture. Based on the accuracy of the first vehicle speed, the accuracy of the second vehicle identification is improved. Attached Figure Description

[0057] Figure 1 : A schematic flowchart of a vehicle violation identification method based on a front-end machine provided in an embodiment of the present invention;

[0058] Figure 2 : A schematic diagram of the structure of a vehicle violation identification system based on a side-mounted machine provided in an embodiment of the present invention;

[0059] Figure 3 : A schematic flowchart of a vehicle violation identification method based on a side-mounted machine, provided for another embodiment of the present invention;

[0060] Figure 4 : A schematic diagram of a second vehicle located in the same lane, provided for another embodiment of the present invention;

[0061] Figure 5 This is a schematic diagram of a second vehicle located in an adjacent lane, as provided in another embodiment of the present invention. Detailed Implementation

[0062] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0063] Example 1

[0064] This invention provides a vehicle violation identification method based on a mobile terminal, the main process of which is described in the following embodiment. Figure 1 It mainly includes steps 101 to 106, each of which specifically includes:

[0065] Step 101: Obtain the first vehicle speed and several first images of the road where the first vehicle is located by using a pre-installed peripheral device in the first vehicle.

[0066] In this embodiment, the step specifically includes: detecting and obtaining the first vehicle speed of the first vehicle through the satellite inertial navigation and inertial measurement unit preset in the device; and capturing several first images of the road where the first vehicle is located through the ADAS camera preset in the device.

[0067] Step 102: Obtain information about the second vehicle based on the first image.

[0068] In this embodiment, the step specifically includes: analyzing the first image through the ADAS camera to obtain information about the second vehicle; the information about the second vehicle includes the lane in which the second vehicle is located, the type of the second vehicle, the license plate, and the relative position of the second vehicle relative to the first vehicle; the lane includes the first lane in which the first vehicle is located and the second lane adjacent to the first lane.

[0069] Step 103: Based on the first vehicle speed and the information of the second vehicle, obtain the second vehicle speed of the second vehicle using a preset speed calculation formula, and compare it with the first vehicle speed to determine whether the second vehicle has committed a violation.

[0070] In this embodiment, the step specifically includes: determining whether the second vehicle is in the first lane or the second lane based on the lane in which the second vehicle is located; if the second vehicle is in the first lane, obtaining a first distance between the first vehicle and the second vehicle through the ADAS camera, and processing the first distance using a preset first speed formula to obtain a second speed of the second vehicle; if the second vehicle is in the second lane, obtaining a second distance between the first vehicle and the second vehicle and a third distance between the first vehicle and the second vehicle from the lane line through the ADAS camera, and processing the second distance and the third distance using a preset second speed formula to obtain a second speed of the second vehicle.

[0071] In this embodiment, after obtaining the second speed of the second vehicle, the second vehicle is compared with the speed limit corresponding to its type to determine whether it is speeding. If the second speed of the second vehicle is greater than or equal to the speed limit, it is determined that the second vehicle is speeding. The relative speed of the second vehicle is obtained based on the first speed and the second speed, and the relative speed is compared with the first speed to determine whether the second vehicle is illegally parked. If the relative speed is equal to the first speed, it is determined that the second vehicle is illegally parked.

[0072] Step 104: If the second vehicle violates the rules, save multiple video trajectories corresponding to several second images where the second vehicle is located, and use a preset vehicle re-identification method to sequentially identify whether the second vehicle in the several second images is the same vehicle.

[0073] In this embodiment, the step specifically includes: performing fine segmentation and secondary analysis on the second image to determine whether the second vehicle in the second image is driving illegally; if the second vehicle is driving illegally, merging multiple video tracks corresponding to the second image, and simultaneously identifying the second vehicle using a preset vehicle re-identification technology based on the license plate and features of the second vehicle in the plurality of second images to determine whether the second vehicle is the same vehicle.

[0074] Step 105: If the second vehicle in the second image is the same vehicle, then the merged video is obtained by comparing the time interval and distance between the multiple video trajectories with a preset trajectory threshold.

[0075] In this embodiment, the step specifically includes: if it is determined that the second vehicle is the same vehicle, then by comparing the time interval between the several video tracks corresponding to the second image with a preset time threshold and comparing the distance between the several video tracks with a preset distance threshold, it is determined whether the second image should be merged; if the time interval between the several video tracks does not exceed the time threshold and the distance does not exceed the distance threshold, then the several second images are merged to obtain a merged video; if the time interval between the several video tracks exceeds the time threshold or the distance exceeds the distance threshold, then the several second images are not merged.

[0076] Step 106: Identify the merged video using a preset violation information identification method to obtain the violation information of the second vehicle.

[0077] In this embodiment, the step specifically includes: obtaining the video trajectory radius corresponding to the merged video by wrapping the merged video; comparing the video trajectory radius with a preset radius threshold to determine whether the merged video is a speeding video or a parking violation video, and obtaining the violation information of the second vehicle based on the merged video.

[0078] On the other hand, this embodiment of the present invention also provides a vehicle violation identification system based on a side-by-side machine, including an acquisition module 201, an information module 202, a judgment module 203, an identification module 204, a merging module 205, and an output module 206.

[0079] The acquisition module 201 is used to acquire the first vehicle speed and several first images of the road where the first vehicle is located through a pre-installed peripheral device in the first vehicle.

[0080] The information module 202 is used to obtain information about the second vehicle based on the first image.

[0081] The judgment module 203 is used to obtain the second speed of the second vehicle based on the first vehicle speed and the information of the second vehicle through a preset speed calculation formula, and compare it with the first vehicle speed to determine whether the second vehicle has violated the rules.

[0082] The identification module 204 is used to save multiple video trajectories corresponding to several second images where the second vehicle is located if the second vehicle violates the rules, and to identify whether the second vehicle in the several second images is the same vehicle in turn by a preset vehicle re-identification method.

[0083] The merging module 205 is used to obtain merged video recordings by comparing the time interval and distance between the multiple video trajectories with a preset trajectory threshold if the second vehicle in the second image is the same vehicle.

[0084] The output module 206 is used to identify the merged video recording using a preset violation information identification method to obtain the violation information of the second vehicle.

[0085] In this embodiment, the acquisition module 201 includes a vehicle speed unit and an image unit;

[0086] The vehicle speed unit is used to detect and obtain the first vehicle speed of the first vehicle through the satellite inertial navigation and inertial measurement unit preset in the device.

[0087] The image unit is used to capture several first images of the road where the first vehicle is located through a pre-set ADAS camera in the device.

[0088] In this embodiment, the judgment module 203 further includes a second vehicle speed unit and a judgment unit.

[0089] The second vehicle speed unit is used to determine whether the second vehicle is in the first lane or the second lane based on its lane location. If the second vehicle is in the first lane, the first distance between the first and second vehicles is obtained through the ADAS camera, and the first distance is processed using a preset first speed formula to obtain the second vehicle speed. If the second vehicle is in the second lane, the second distance between the first and second vehicles and the lane line and the third distance between the first and second vehicles are obtained through the ADAS camera, and the second and third distances are processed using a preset second speed formula to obtain the second vehicle speed. The judgment unit is used to determine whether the second vehicle is speeding based on its second vehicle speed and type, by comparing it with the speed limit corresponding to the type of the second vehicle. If the second vehicle speed is greater than or equal to the speed limit, the second vehicle is determined to be speeding. Based on the first and second speeds, the relative speed of the second vehicle is obtained, and by comparing the relative speed with the first speed, the second vehicle is determined to be illegally parked. If the relative speed is equal to the first speed, the second vehicle is determined to be illegally parked.

[0090] In this embodiment, the identification module 204 further includes a secondary analysis unit and a re-identification unit.

[0091] The secondary analysis unit is used to determine whether the second vehicle in the second image is driving illegally by performing fine segmentation and secondary analysis on the second image.

[0092] The re-identification unit is used to merge multiple video trajectories corresponding to the second image if the second vehicle has committed a traffic violation. At the same time, based on the license plate and features of the second vehicle in the plurality of second images, the unit identifies the second vehicle using a preset vehicle re-identification technology to determine whether the second vehicle is the same vehicle.

[0093] This embodiment provides a vehicle violation identification method and system based on a vehicle edge camera. By installing a vehicle edge camera on a first vehicle, the system acquires the first vehicle's speed and the specific road conditions where the first vehicle is currently located. Due to the stable nature of the edge camera, it can accurately and promptly obtain the vehicle's speed and road conditions, improving the stability of violation identification. Then, based on the first image, various information about a second vehicle is obtained for subsequent identification. After obtaining the speed of the first vehicle and the lane and relative position of the second vehicle, a preset speed calculation formula is used to process the lane and relative position, allowing the speed of the second vehicle to be obtained according to different lanes and distances, improving the accuracy of the second vehicle speed calculation and further enhancing the accuracy of violation identification. The second vehicle's second speed and type are obtained, and compared with the speed limit corresponding to the first speed and the type to determine whether the second vehicle has committed illegal parking or speeding. The vehicle is classified to avoid errors in violation identification and improve the accuracy of identification. At the same time, the stability of the first speed measurement ensures the accuracy of the judgment, thereby improving the accuracy of identification. When it is determined that the second vehicle in the first image has committed a violation, the images of the violation are merged, and then the merged video is identified to determine whether the second vehicle in the merged violation images is the same vehicle. Then, the second images of the same vehicle are merged to obtain a merged video, and the violation information of the second vehicle that committed the violation is obtained from the merged video, thus obtaining the complete violation information of the second vehicle.

[0094] Example 2

[0095] This embodiment provides another method for vehicle violation identification based on a mobile terminal unit. The main process is described in [reference needed]. Figure 3 It mainly includes steps 301 to 306, each step mainly includes:

[0096] Step 301: Determine the vehicle speed and the first image of the road where the vehicle is located based on the preset side machine in the vehicle.

[0097] In this embodiment, the step specifically includes: detecting and obtaining the first vehicle speed of the first vehicle through the satellite inertial navigation and inertial measurement unit preset in the device; and capturing several first images of the road where the first vehicle is located through the ADAS camera preset in the device.

[0098] In this embodiment, the first vehicle is the self-driving vehicle. First, the vehicle speed is determined by the function of the standard unit. Under normal circumstances, the vehicle speed is given by the satellite inertial navigation system. When driving in a tunnel, the vehicle speed is determined by the satellite navigation system and the IMU (Inertial Measurement Unit). At the same time, the ADAS camera included in the standard unit captures the road conditions around the self-driving vehicle.

[0099] Step 302: Calculate the speed of the other car in the first image.

[0100] In this embodiment, the step specifically includes: analyzing the first image through the ADAS camera to obtain information about the second vehicle, determining whether the second vehicle is in the first lane or the second lane based on the lane in which the second vehicle is located, and then obtaining the speed of the second vehicle based on the first speed of the first vehicle.

[0101] In this embodiment, the second vehicle is the other vehicle, and the second vehicle speed is the speed of the other vehicle. The lane of the other vehicle is determined by the ADAS function of the vehicle's speedometer, and then the relative speed of the other vehicle is calculated. In this embodiment, it is first determined whether the lane of the other vehicle is the driver's lane or an adjacent lane. When the other vehicle is in front of the driver's lane, please refer to... Figure 4 This is a schematic diagram illustrating a situation where another vehicle is located in the vehicle's lane, as provided in an embodiment of the present invention. Figure 4 In the above scenario, assuming that at time t0, the vehicle's speed is v0 (m / s) and the distance to the vehicle in front is x0 meters, and at time t1, the vehicle's speed is v1 (m / s) and the distance to the vehicle in front is x1 meters, then the speed of the other vehicle can be obtained using the average speed formula described below. The average speed formula is as follows:

[0102]

[0103] In this embodiment, when the other vehicle is located in other adjacent lanes, the ADAS camera first determines whether the other vehicle's lane is within several adjacent lanes. Please refer to [reference needed]. Figure 5 This is a schematic diagram of a vehicle in an adjacent lane provided by an embodiment of the present invention. Let the distance between the lane lines be d meters. At time t0, the relative distance between the other vehicle and the vehicle itself is d0 meters, and the vehicle's speed is v0. At time t1, the relative distance between the other vehicle and the vehicle itself is d1 meters, and the vehicle's speed is v1. The speed of the other vehicle can be obtained using the average speed formula described below. The average speed formula is:

[0104]

[0105] Step 303: Based on the speed of the other vehicle, identify the other vehicle as having committed a traffic violation and save the violation image.

[0106] In this embodiment, the step mainly includes: determining whether the second vehicle violates the rules by comparing the second vehicle's second speed and type with the first speed corresponding to the second vehicle's type; and if the second vehicle violates the rules, saving multiple video trajectories corresponding to several second images of the second vehicle.

[0107] In this embodiment, ADAS is used for vehicle detection, outputting the vehicle category, which includes at least two types: passenger cars and non-passenger cars. When the second vehicle speed is greater than the speed limit corresponding to the second vehicle type, that is, when the average speed of the other vehicle exceeds the speed limit of the vehicle type, it is judged as speeding; when the relative speed of the other vehicle is equal to or less than the speed of the vehicle itself, it is judged as illegal parking; when the relative speed is less than or less than the speed of the vehicle itself, it is judged as reversing. To enhance robustness, the average relative speed of the other vehicle during the time it passes by can be calculated, and the average relative speed can be used for judgment. At the same time, the image of the other vehicle where the violation occurred is saved and sent to the cloud platform.

[0108] Step 304: Perform secondary analysis on the violation image and determine whether the other vehicles in the violation image belong to the same vehicle.

[0109] In this embodiment, the step specifically includes: performing fine segmentation and secondary analysis on the second image to determine whether the second vehicle in the second image is driving illegally; and simultaneously using a preset vehicle re-identification method to identify the second vehicle in the second image to determine whether the second vehicle is the same vehicle.

[0110] In this embodiment, when the cloud platform receives the sequence of video image frames of another vehicle that has committed a traffic violation, it performs fine segmentation of the road surface to more accurately determine whether there is a traffic violation in the captured video. It also connects the received violation trajectory segments to form a continuous trajectory that is as long as possible. At the same time, based on the license plate of the other vehicle, i.e., the vehicle characteristics of the other vehicle, it uses a preset vehicle ReID (Re-identification) method to identify whether the other vehicle is the same vehicle.

[0111] Step 305: Merge traffic violation images of the same vehicle and determine whether the traffic violation images belong to the same trip.

[0112] In this embodiment, the step mainly includes: if it is determined that the second vehicle is the same vehicle, then by comparing the time interval between the several video tracks corresponding to the second image with a preset time threshold and comparing the distance between the several video tracks with a preset distance threshold, it is determined whether the second image should be merged.

[0113] In this embodiment, after determining that the other vehicles in the violation images belong to the same vehicle, several violation images containing the same vehicle are merged. Then, the distance and time interval between the several violation images are calculated. First, the distance is compared with a preset distance threshold. If the distance does not exceed the distance threshold, then the time interval is compared with a preset time threshold. If the time interval does not exceed the time threshold, then the violation images are merged.

[0114] Step 306: Merge traffic violation images belonging to the same trip, identify the merged violation images, determine and output the violation information.

[0115] In this embodiment, the step mainly includes: obtaining the video trajectory radius corresponding to the merged video by wrapping the merged video; comparing the video trajectory radius with a preset radius threshold to determine whether the merged video is a speeding video or a parking violation video, and obtaining the violation information of the second vehicle based on the merged video.

[0116] In this embodiment, after merging violation images belonging to the same trip, the trajectory corresponding to the violation image is wrapped, and then the radius of the wrapped trajectory is calculated and compared with a preset trajectory radius threshold to determine whether the violation trajectory corresponds to illegal parking or speeding. Then, the violation information contained in the trajectory corresponding to the violation image is output, including more macro-level violation information such as speeding section / duration and illegal parking location / duration.

[0117] This embodiment provides another vehicle violation identification method based on a mobile terminal unit (MCU). By utilizing the MCU's ADAS and GPS / INS functions, it can dynamically capture speeding and illegally parked vehicles on highways, overcoming the limitations of fixed-location capture. In tunnels, the MCU's GPS / INS and IMU can be combined to calculate vehicle speed, thereby capturing violations. The transmitted violation video images are then analyzed a second time using a cloud platform. After further identifying the violation, the violation segments are merged into a long, continuous trajectory to output more comprehensive violation information.

[0118] The specific embodiments described above further illustrate the purpose, technical solution, and beneficial effects of the present invention. It should be understood that the above descriptions are merely specific embodiments of the present invention and are not intended to limit the scope of protection of the present invention. In particular, it should be noted that any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention for those skilled in the art.

Claims

1. A vehicle violation identification method based on a mobile terminal, characterized in that, include: The first vehicle speed and several first images of the road where the first vehicle is located are obtained by a pre-installed device in the first vehicle. Based on the first image, obtain information about the second vehicle; Based on the first vehicle speed and the information of the second vehicle, a second vehicle speed is obtained using a preset speed calculation formula, and compared with the first vehicle speed to determine whether the second vehicle has violated the rules. Specifically, this includes: determining whether the second vehicle is in the first lane or the second lane based on the lane in which the second vehicle is located, where the lane includes the first lane in which the first vehicle is located and the second lane adjacent to the first vehicle; if the second vehicle is in the first lane, a first distance between the first vehicle and the second vehicle is obtained through an ADAS camera, and the first distance is processed using a preset first speed formula to obtain the second vehicle speed; if the second vehicle is in the second lane, a second distance between the first vehicle and the lane line and a third distance between the first vehicle and the second vehicle are obtained through the ADAS camera, and the second distance and the third distance are processed using a preset second speed formula to obtain the second vehicle speed. If the second vehicle violates the rules, multiple video trajectories corresponding to several second images containing the second vehicle are saved, and the second vehicle in the several second images is identified as the same vehicle in turn by a preset vehicle re-identification method. If the second vehicle in the second image is the same vehicle, then the merged video is obtained by comparing the time interval and distance between the multiple video trajectories with a preset trajectory threshold. The merged video is identified using a preset violation information identification method to obtain the violation information of the second vehicle.

2. The vehicle violation identification method based on a mobile terminal as described in claim 1, characterized in that, The step of acquiring the first vehicle's first speed and several first images of the road where the first vehicle is located through a pre-installed peripheral device in the first vehicle specifically includes: The first vehicle speed is obtained by detecting the satellite inertial navigation and inertial measurement units preset in the device. The device uses an ADAS camera pre-installed in the device to capture several first images of the road where the first vehicle is located.

3. The vehicle violation identification method based on a mobile terminal as described in claim 2, characterized in that, The step of obtaining information about the second vehicle based on the first image specifically includes: The ADAS camera analyzes the first image to obtain information about the second vehicle. The information about the second vehicle includes the lane in which the second vehicle is located, the type of the second vehicle, the license plate, and the relative position of the second vehicle to the first vehicle. The lane includes the first lane in which the first vehicle is located and the second lane adjacent to the first vehicle.

4. The vehicle violation identification method based on a mobile terminal as described in claim 1, characterized in that, The step of comparing the second vehicle speed with the first vehicle speed to determine whether the second vehicle has committed a violation specifically includes: Based on the second vehicle's second speed and type, a comparison is made with the speed limit corresponding to the type of the second vehicle to determine whether the second vehicle is speeding. If the second vehicle's second speed is greater than or equal to the speed limit, then the second vehicle is determined to be speeding. Based on the first vehicle speed and the second vehicle speed, the relative speed of the second vehicle is obtained, and by comparing the relative speed with the first vehicle speed, it is determined whether the second vehicle has illegally parked. If the relative speed is equal to the first vehicle speed, then it is determined that the second vehicle has illegally parked.

5. The vehicle violation identification method based on a mobile terminal as described in claim 1, characterized in that, The step of sequentially identifying whether the second vehicles in the plurality of second images are the same vehicle using a preset vehicle re-identification method specifically includes: By performing fine segmentation and secondary analysis on the second image, it is determined whether the second vehicle in the second image is driving illegally; If the second vehicle is found to be driving illegally, the multiple video tracks corresponding to the second image are merged. At the same time, based on the license plate and features of the second vehicle in the multiple second images, the second vehicle is identified using a preset vehicle re-identification technology to determine whether the second vehicle is the same vehicle.

6. The vehicle violation identification method based on a mobile terminal as described in claim 1, characterized in that, The step of comparing the time intervals and distances between the multiple video tracks with a preset track threshold to obtain merged video recordings specifically includes: If it is determined that the second vehicle is the same vehicle, then the second image is merged by comparing the time interval between the several video tracks corresponding to the second image with a preset time threshold and comparing the distance between the several video tracks with a preset distance threshold. If the time interval between the plurality of video tracks does not exceed the time threshold and the distance between them does not exceed the distance threshold, then the plurality of second images are merged to obtain a merged video. If the time interval between the plurality of video tracks exceeds the time threshold or the distance between them exceeds the distance threshold, then the plurality of second images will not be merged.

7. The vehicle violation identification method based on a mobile terminal as described in claim 1, characterized in that, The step of identifying the merged video recording using a preset violation identification method to obtain the violation information of the merged video recording specifically includes: The radius of the video trajectory corresponding to the merged video is obtained by wrapping the merged video. Based on the radius of the video trajectory, it is compared with a preset radius threshold to determine whether the merged video is a speeding video or a parking violation video, and at the same time, the violation information of the second vehicle is obtained based on the merged video.

8. A vehicle violation identification system based on a mobile terminal, characterized in that, It includes an acquisition module, an information module, a judgment module, an identification module, a merging module, and an output module; The acquisition module is used to acquire the first vehicle speed and several first images of the road where the first vehicle is located through a pre-installed peripheral device in the first vehicle. The information module is used to obtain information about the second vehicle based on the first image; The judgment module is used to obtain the second vehicle speed of the second vehicle based on the first vehicle speed and the information of the second vehicle, using a preset speed calculation formula, and compare it with the first vehicle speed to determine whether the second vehicle has violated the rules. Specifically, it includes: determining whether the second vehicle is in the first lane or the second lane based on the lane where the second vehicle is located, wherein the lane includes the first lane where the first vehicle is located and the second lane adjacent to the first vehicle; if the second vehicle is in the first lane, obtaining the first distance between the first vehicle and the second vehicle through the ADAS camera, processing the first distance through a preset first speed formula to obtain the second vehicle speed; if the second vehicle is in the second lane, obtaining the second distance between the first vehicle and the second vehicle and the lane line and the third distance between the first vehicle and the second vehicle through the ADAS camera, processing the second distance and the third distance through a preset second speed formula to obtain the second vehicle speed. The identification module is used to save multiple video trajectories corresponding to several second images where the second vehicle is located if the second vehicle violates the rules, and to identify whether the second vehicle in the several second images is the same vehicle in turn through a preset vehicle re-identification method. The merging module is used to obtain merged video recordings by comparing the time interval and distance between the multiple video trajectories with a preset trajectory threshold if the second vehicle in the second image is the same vehicle. The output module is used to identify the merged video recording using a preset violation information identification method to obtain the violation information of the second vehicle.

9. A vehicle violation identification system based on a mobile terminal as described in claim 8, characterized in that, The acquisition module includes a vehicle speed unit and an image unit; The vehicle speed unit is used to detect and obtain the first vehicle speed of the first vehicle through the satellite inertial navigation and inertial measurement unit preset in the device. The image unit is used to capture several first images of the road where the first vehicle is located through a pre-set ADAS camera in the device.

Images