Road camera video selection method and system based on camera dynamic weight

Abstract

The invention provides a road camera video selection method and system based on a camera dynamic weight, and the method comprises the steps: building a rectangular region with the longitude and latitude of a vehicle as a center, and dividing the rectangular region into four sub-regions; determining a driving scene of the vehicle by using a difference value between the heading angles of the vehicle at the required moment and the previous moment; adjusting the area weight of the at least one sub-area according to the driving scene of the vehicle; determining the priority of each roadside camera by using the driving scene weight of each roadside camera, the area weight of the sub-area where each roadside camera is located and the current shared scene weight of each roadside camera; according to the sequence of the priorities from high to low, the videos of the roadside cameras with the preset number are selected, so that a camera acquisition algorithm better fits video selection tendencies under different road conditions and different vehicle running states, the problem that a fixed algorithm cannot flexibly control selection expectation is avoided, and the accuracy of video selection is improved. And the monitoring accuracy and the response efficiency in a complex road network environment are improved.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent transportation, and in particular, to a method and system for selecting road camera videos based on dynamic weights of cameras. Background Art

[0002] With the gradual expansion of vehicle autonomous driving demonstration areas and the construction and improvement of autonomous driving supervision systems, quickly obtaining videos of abnormal driving and accidents of fully unmanned autonomous vehicles has become a technical difficulty in the supervision of autonomous vehicles. When abnormal driving or an accident occurs to an autonomous vehicle, how to accurately and quickly obtain roadside videos as supplementary evidence materials is an urgent problem to be solved in the automatic monitoring and supervision industry.

[0003] The currently mainstream roadside video acquisition solutions generally adopt the method of manually retrieving camera videos based on the location and time of the incident, or obtaining camera videos based on the prediction of the vehicle's driving trajectory, direction, time, and camera shooting range.

[0004] Currently, the above methods all have obvious defects: The method of manually screening cameras and recording videos has a high accuracy of video acquisition, but high labor costs and time costs, and low efficiency, which cannot meet the current situation of an increasing number of supervised vehicles. The automated video acquisition method of obtaining videos based on the vehicle's driving trajectory direction and camera shooting range also has problems. Limited by the fixed camera selection logic, it cannot adapt to complex and changeable road conditions and camera selection tendencies in different scenarios, resulting in the accuracy of obtaining camera videos not meeting the supervision requirements.

[0005] Therefore, how to improve the monitoring accuracy and response efficiency in a complex road network environment is a technical problem that needs to be urgently solved by those skilled in the art. Summary of the Invention

[0006] The present invention provides a method and system for selecting road camera videos based on dynamic weights of cameras to solve the defects of low monitoring accuracy and low response efficiency in a complex road network environment in the prior art.

[0007] On the one hand, the present invention provides a method for selecting road camera videos based on dynamic weights of cameras, which includes: Construct a rectangular area centered on the longitude and latitude of the vehicle, and divide the rectangular area into 4 sub-areas; wherein, the long axis direction of the rectangular area is consistent with the vehicle driving direction, and the mapped longitudes and latitudes of multiple roadside cameras are included in the rectangular area; Use the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment of the required moment to determine the driving scenario of the vehicle; Adjust the regional weights of at least one sub-region according to the driving scenario of the vehicle; Determine the priority of each roadside camera by using the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the current shared scenario weight of each roadside camera; Select the videos of a preset number of roadside cameras in the order of decreasing priority.

[0008] According to a method for selecting road camera videos based on dynamic weights of cameras provided by the present invention, use the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment of the required moment to determine the driving scenario of the vehicle, including: If the absolute value of the difference is less than or equal to a preset absolute value, determine the driving scenario of the vehicle according to the scenario range value to which the difference belongs; If the absolute value of the difference is greater than the preset absolute value, sum or subtract the difference from a preset value to obtain a conversion value corresponding to the difference, and determine the driving scenario of the vehicle according to the scenario range value to which the conversion value belongs.

[0009] According to a method for selecting road camera videos based on dynamic weights of cameras provided by the present invention, if the target value is greater than a first preset difference of the scenario range value and less than a second preset difference of the scenario range value, determine that the driving scenario of the vehicle is a right turn; wherein, the first preset difference is a positive number; wherein, the target value includes the difference or the conversion value; If the target value is greater than or equal to a third preset difference of the scenario range value and less than or equal to the first preset difference, determine that the driving scenario of the vehicle is a straight line; wherein, the third preset difference is a negative number; If the target value is greater than a fourth preset difference of the scenario range value and less than the third preset difference, determine that the driving scenario of the vehicle is a left turn; wherein, the preset absolute value is the difference between the second preset difference and the fourth preset difference.

[0010] According to a method for selecting road camera videos based on dynamic weights of cameras provided by the present invention, adjusting the regional weights of at least one sub-region according to the driving scenario of the vehicle includes: According to the driving scenario of the vehicle, use the overlap degree between the shooting range of each roadside camera and the vehicle trajectory to predict the predicted shooting duration of the vehicle in each sub-region; Select the at least one sub-region in the order of decreasing predicted shooting duration, and adjust the regional weights of the at least one sub-region.

[0011] A method for selecting road camera videos based on dynamic weights of cameras according to the present invention further includes: If the driving scenario of the vehicle is a left turn or a right turn, expand the width of the at least one sub-region by a preset multiple, or If the driving scenario of the vehicle is a left turn or a right turn, obtain the turning radius of the vehicle, determine an expansion coefficient based on a preset correlation between the turning radius and the expansion coefficient, and expand the width of the at least one sub-region according to the expansion coefficient.

[0012] A method for selecting road camera videos based on dynamic weights of cameras according to the present invention determines the priority of each roadside camera by using the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the shared scenario weight of each roadside camera, including: Calculate the product of the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the shared scenario weight of each roadside camera; Set the priority of each roadside camera in descending order of the product.

[0013] A method for selecting road camera videos based on dynamic weights of cameras according to the present invention further includes: If the weight of any target camera among the remaining cameras in the camera cluster to which the i-th roadside camera among the multiple roadside cameras belongs changes, obtain the adjusted weight of the any target camera; wherein, the camera cluster is obtained by clustering all the roadside cameras to be processed; Use the adjusted weight to determine the change amount of the shared scenario weight of the i-th roadside camera; Take the sum of the original shared scenario weight of the i-th roadside camera and the change amount as the current shared scenario weight of the i-th roadside camera.

[0014] A method for selecting road camera videos based on dynamic weights of cameras according to the present invention determines the change amount of the shared scenario weight of the i-th roadside camera by using the adjusted weight, including: Determine the similarity between the i-th roadside camera and the target camera; Determine the weight transfer coefficient corresponding to the i-th roadside camera according to a preset correlation between the similarity and the weight transfer coefficient; Take the product of the adjusted weight and the weight transfer coefficient as the change amount.

[0015] On the other hand, the present invention also provides a system for selecting road camera videos based on dynamic weights of cameras, which includes: A building module for building a rectangular area centered on the longitude and latitude of a vehicle and dividing the rectangular area into 4 sub-areas; wherein, the major axis direction of the rectangular area is consistent with the vehicle driving direction, and the longitude and latitude of multiple roadside cameras are included in the rectangular area; A first determination module for determining the driving scenario of the vehicle by using the difference between the first heading angle of the vehicle at the demand moment and the second heading angle of the vehicle at the previous moment of the demand moment; An adjustment module for adjusting the area weight of each sub-area according to the driving scenario of the vehicle; A second determination module for determining the priority of each roadside camera by using the driving scenario weight of each roadside camera, the area weight of the sub-area where each roadside camera is located, and the current shared scenario weight of each roadside camera; A selection module for selecting videos of a preset number of roadside cameras in the order from high to low priority.

[0016] On the other hand, the present invention also provides an electronic device, which includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the method for selecting road camera videos based on dynamic camera weights as described in any one of the above.

[0017] On the other hand, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the method for selecting road camera videos based on dynamic camera weights as described in any one of the above.

[0018] On the other hand, the present invention also provides a computer program product, including a computer program. When the computer program is executed by a processor, it implements the method for selecting road camera videos based on dynamic camera weights as described in any one of the above.

[0019] The method and system for selecting road camera videos based on dynamic weights of cameras provided by the present invention construct a rectangular area centered on the longitude and latitude of a vehicle, and divide the rectangular area into 4 sub-areas; use the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment of the required moment to determine the driving scenario of the vehicle; adjust the area weights of at least one sub-area according to the driving scenario of the vehicle; use the driving scenario weights of each roadside camera, the area weights of the sub-areas where each roadside camera is located, and the current shared scenario weights of each roadside camera to determine the priority of each roadside camera; select videos of a preset number of roadside cameras in the order of decreasing priority. In this way, the camera acquisition algorithm can be made more suitable for the video selection tendencies under different road conditions and different vehicle operating states, avoiding the problem that a fixed algorithm cannot flexibly control the selection as expected, and improving the monitoring accuracy and response efficiency in a complex road network environment. BRIEF DESCRIPTION OF THE DRAWINGS

[0020] In order to more clearly illustrate the technical solutions in the present invention or the prior art, the following will briefly introduce the drawings required for the description of the embodiments or the prior art. Obviously, the drawings in the following description are some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0021] Figure 1 is a flowchart of the method for selecting road camera videos based on dynamic weights of cameras provided by an embodiment of the present invention; Figure 2 is a schematic diagram of a rectangular area constructed centered on the longitude and latitude of a vehicle; Figure 3 is a schematic diagram of sub-areas C and D that need to have their weights adjusted in a straight-ahead scenario; Figure 4 is a schematic diagram of sub-areas C and D that need to have their weights adjusted in a left-turn scenario; Figure 5 is a schematic diagram of sub-areas A and C that need to have their weights adjusted in a right-turn scenario; Figure 6 is a schematic diagram of the structure of the system for selecting road camera videos based on dynamic weights of cameras provided by an embodiment of the present invention; Figure 7 is a schematic diagram of the structure of an electronic device provided by an embodiment of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0022] To make the objectives, technical solutions, and advantages of the present invention clearer, the technical solutions in the present invention will be clearly and completely described below with reference to the accompanying drawings in the present invention. Obviously, the described embodiments are some, but not all, of the embodiments of the present invention. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in the present invention without creative efforts shall fall within the protection scope of the present invention.

[0023] Figure 1 It is a schematic flowchart of a method for selecting road camera videos based on dynamic weights of cameras according to an embodiment of the present invention.

[0024] As Figure 1 shown, the execution subject of the method for selecting road camera videos based on dynamic weights of cameras according to an embodiment of the present invention may be an electronic device. The method mainly includes the following steps: 101. Construct a rectangular area centered on the longitude and latitude of the vehicle, and divide the rectangular area into 4 sub-areas; In a specific implementation process, the running data reported by an autonomous vehicle can be used to obtain the longitude and latitude information and the heading angle information of the vehicle. A rectangular area can be constructed centered on the longitude and latitude of the vehicle, and the rectangular area can be divided into 4 sub-areas, as Figure 2 shown. Figure 2 It is a schematic diagram of a rectangular area constructed centered on the longitude and latitude of the vehicle. The rectangular area may include four sub-areas A to D. Among them, the long axis direction of the rectangular area is consistent with the vehicle driving direction. The rectangular area contains the mapped longitudes and latitudes of multiple roadside cameras. That is to say, after projection, the projection coordinates of multiple roadside cameras can fall within the rectangular area.

[0025] It should be noted that the width of the rectangular area may be the same as the width of the current driving road and change with the change of the width of the current driving road. The length of the rectangular area can be set according to the driving speed of the vehicle. Among them, a first speed threshold and a second speed threshold can be set, and the second speed threshold is greater than the first speed threshold. When the driving speed of the vehicle is less than or equal to the first speed threshold, the default length is adopted. When the driving speed of the vehicle is greater than the first speed threshold but less than the second speed threshold, the ratio of the driving speed of the vehicle to the first speed threshold can be calculated, and the product of the ratio and the default length can be calculated to obtain the length of the rectangular area. That is, L1 = L0 * a, where L1 is the length of the rectangular area, L0 is the default length, and a is the ratio of the driving speed of the vehicle to the speed threshold. When the driving speed of the vehicle is greater than the second speed threshold, the length of the rectangular area is 2L0.

[0026] 102. Determine the driving scenario of the vehicle by using the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment of the required moment; In a specific implementation process, the first heading angle of the vehicle at the required moment and the difference between the second heading angle of the vehicle at the previous moment of the required moment can be obtained from the heading angle information of the vehicle, and the difference between the first heading angle and the second heading angle can be calculated. Further, based on this difference, the driving scenario of the vehicle can be determined. Among them, the driving scenario may include going straight, turning left, and turning right. The required moment is the recording moment of the event to be viewed, such as the accident occurrence moment, etc.

[0027] In a specific implementation process, respective scenario range values can be set according to different driving scenarios of the vehicle. After obtaining the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment, first judge the magnitude relationship between this difference and the preset absolute value. If the absolute value of the difference is less than or equal to the preset absolute value, the difference can be directly used, and the scenario range value to which the difference belongs can be determined, and then the driving scenario of the vehicle can be determined. If the absolute value of the difference is greater than the preset absolute value, the difference cannot be directly used to determine the driving scenario of the vehicle. Instead, the difference can be added to or subtracted from the preset value to obtain the conversion value corresponding to the difference, and the driving scenario of the vehicle can be determined according to the scenario range value to which the conversion value belongs.

[0028] Specifically, if the target value is greater than the first preset difference of the scenario range value and less than the second preset difference of the scenario range value, determine that the driving scenario of the vehicle is turning right; among them, the first preset difference is a positive number; among them, the target value includes the difference or the conversion value; If the target value is greater than or equal to the third preset difference of the scenario range value and less than or equal to the first preset difference, determine that the driving scenario of the vehicle is going straight; among them, the third preset difference is a negative number; If the target value is greater than the fourth preset difference of the scenario range value and less than the third preset difference, determine that the driving scenario of the vehicle is turning left; among them, the preset absolute value is the difference between the second preset difference and the fourth preset difference.

[0029] Specifically, taking the first preset difference as 20°, the second preset difference as 120°, the third preset difference as -20°, the fourth preset difference as -120°, and the preset absolute value as 240° as an example, the respective scenario range values for different driving scenarios are as follows: The change in the heading angle (the difference between the heading angles at the front and rear moments) is greater than 20° and less than 120°, and it is determined that the driving scenario of the vehicle is turning right.

[0030] When the change in the heading angle is greater than -20° and less than 20°, it is determined that the driving scenario of the vehicle is going straight.

[0031] When the change in the heading angle is greater than -120° and less than -20°, it is determined that the driving scenario of the vehicle is turning left.

[0032] When the absolute value of the change in the heading angle is greater than 240°, special processing needs to be performed by adding or subtracting 360° before determining the driving scenario of the vehicle. For example, if the heading angle at the required moment is 10° and the heading angle at the previous moment is 280°, the difference between the two is -270°. At this time, adding 360° to this difference, it is determined that the driving scenario of the vehicle is turning right. If the heading angle at the required moment is 280° and the heading angle at the previous moment is 10°, the difference between the two is 270°. At this time, subtracting 360° from this difference, it is determined that the driving scenario of the vehicle is turning left.

[0033] It should be noted that the determination range of the change in the heading angle can be adjusted through dynamic parameter configuration to optimize the accuracy of determining the driving direction of the vehicle. For example, if the noise of the heading angle sensor is large, the straight-ahead interval can be expanded to filter out interference; if the accuracy is high, the interval can be reduced to improve sensitivity, and no more examples will be given here.

[0034] 103. Adjust the regional weights of at least one sub-region according to the driving scenario of the vehicle; In a specific implementation process, according to the driving scenario of the vehicle, the overlap degree between the shooting range of each roadside camera and the vehicle trajectory can be used to predict the predicted shooting duration of the vehicle in each sub-region; select the at least one sub-region in the order of the predicted shooting duration from high to low, and adjust the regional weights of the at least one sub-region, and the length or width of at least one sub-region can be adjusted. That is to say, preferentially select the sub-regions corresponding to the cameras with a high overlap degree between the shooting range and the vehicle trajectory and that can shoot the vehicle for a long time for weight adjustment and length or width adjustment.

[0035] Specifically, reference can be made to Figures 3 to 5 to determine at least one sub-region that needs to have its weight adjusted. Among them, Figure 3 is a schematic diagram of sub-regions C and D that need to have their weights adjusted in a straight-ahead scenario, Figure 4 is a schematic diagram of sub-regions C and D that need to have their weights adjusted in a left-turn scenario, Figure 5 is a schematic diagram of sub-regions A and C that need to have their weights adjusted in a right-turn scenario.

[0036] Specifically, as Figure 3 shown, Figure 3Among them, (a1)-(a5) are respectively the shooting ranges of the lower-left roadside camera, the lower-left roadside camera, the upper-right roadside camera, the lower-right roadside camera, and the upper-left roadside camera. Among them, each shooting range is represented by two facing triangles, and (a1) and (a2) represent the states at different moments when going straight. From Figure 3 It can be seen that when the vehicle is in a straight state, the video shooting areas of the cameras in this lane (the lower-left roadside camera and the lower-right roadside camera) coincide most with the vehicle's driving trajectory, and the cameras can capture the vehicle for a long time. Therefore, the weights of sub-regions C and D are adjusted to 2, and the initial weights of sub-regions A and B are maintained at 1, or the initial weights of sub-regions A and B can also be reduced to 0.5.

[0037] As Figure 4 shown, Figure 4 Among them, (a1)-(a4) are respectively the shooting ranges of the lower-left roadside camera, the upper-right roadside camera, the upper-left roadside camera, and the lower-right roadside camera. From Figure 4 It can be seen that the video shooting areas of the oncoming lane (the upper-right roadside camera) and the lower lane (the lower-right roadside camera) coincide most with the vehicle's driving trajectory, and the cameras can capture the current vehicle for a long time. Therefore, the weights of sub-regions C and D are adjusted to 2, and the initial weights of sub-regions A and B are maintained at 1, or the initial weights of sub-regions A and B can also be reduced to 0.5. At the same time, the vehicle speed is slower in the turning state, and although the expected selected roadside camera has a higher shooting angle and shooting duration than other roadside cameras, due to the factors of the roadside camera acquisition area, it may cause the expected selected roadside camera not to appear in the camera selection area, resulting in incorrect video selection. Therefore, when it is determined that the vehicle driving scenario is a left turn, the widths of sub-regions C and D can be adjusted to 2 times the initial width, so as to increase the probability that the expected camera can be covered by the camera acquisition area, and the turning radius of the vehicle can also be obtained. Based on the preset correlation between the turning radius and the expansion coefficient, the expansion coefficient is determined, and the widths of the at least one sub-region are expanded according to the expansion coefficient.

[0038] As Figure 5 shown, Figure 5 Among them, (a1)-(a4) are respectively the shooting ranges of the lower-left roadside camera, the upper-left roadside camera, the lower-right roadside camera, and the upper-right roadside camera. From Figure 5It can be seen that the shooting ranges of the upper area (the roadside camera in the upper left corner) and the camera of this lane (the roadside camera in the lower left corner) coincide highly with the vehicle driving trajectory, and the current vehicle can be photographed for a long time. Therefore, the weights of sub-regions A and C are adjusted to 2, while maintaining the initial weights of 1 for sub-regions B and D, or the initial weights of sub-regions B and D can also be reduced to 0.5. At the same time, the vehicle speed is slow in the turning state, and although the expected selected roadside camera has a higher shooting angle and shooting duration than other roadside cameras, due to the factors of the area obtained by the roadside camera, the expected selected roadside camera may not appear in the camera selection area, resulting in an incorrect video selection. Therefore, when it is determined that the vehicle driving scenario is a right turn, the widths of sub-regions A and C can be adjusted to twice the initial width, so as to increase the probability that the expected camera can be covered by the camera acquisition area, and the turning radius of the vehicle can also be obtained. Based on the preset correlation between the turning radius and the expansion coefficient, the expansion coefficient is determined, and the widths of the at least one sub-region are expanded according to the expansion coefficient.

[0039] In this embodiment, the vehicle driving scenario can be determined according to the change of the vehicle heading angle in the vehicle operation data, and the dynamic weight adjustment of the obtained sub-region can be performed, so that the camera acquisition algorithm is more in line with the video selection tendency under different road conditions and different vehicle operation states, and the problem that the fixed algorithm cannot flexibly control the selection expectation is avoided.

[0040] 104. Determine the priority of each roadside camera by using the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the current shared scenario weight of each roadside camera; In a specific implementation process, the product of the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the shared scenario weight of each roadside camera can be calculated; the priorities of each roadside camera are set in the order from large to small according to the product. For example, for the driving scenario of a left turn, the priority of each roadside camera = the weight of the camera for left turn * the shared weight for left turn * the weight of the area where the camera is located (one of the weights of sub-regions A, B, C, and D).

[0041] In a specific implementation process, the shared scenario weight of each roadside camera can be obtained in the following manner: (1) If the weight of any target camera among the remaining cameras in the camera cluster to which the i-th roadside camera in the multiple roadside cameras belongs changes, obtain the adjusted weight of the any target camera; Specifically, all roadside cameras to be processed can be clustered based on information such as the same street, the same camera shooting angle, and the same camera device pole position of the cameras, to obtain multiple camera clusters. It is possible to determine the camera cluster to which the i-th roadside camera among the multiple roadside cameras in the rectangular area belongs. When the weight of any target camera among the other cameras in this camera cluster changes, the adjusted weight of the any target camera is obtained.

[0042] (2) Using the adjusted weight, determine the change amount of the shared scene weight of the i-th roadside camera; After obtaining the adjusted weight of any target camera among the other cameras, the similarity between the i-th roadside camera and the target camera can be determined using a preset similarity algorithm (such as the cosine similarity algorithm). According to the association relationship between the preset similarity and the weight transfer coefficient, the weight transfer coefficient corresponding to the i-th roadside camera is determined, and the product of the adjusted weight and the weight transfer coefficient is used as the change amount.

[0043] That is to say, the roadside cameras in the same camera cluster usually have approximate features. If the shared scene weight of a roadside camera changes, it usually causes other roadside cameras to change accordingly. However, it does not change directly according to the adjusted weight of the roadside camera whose weight has changed. Instead, it is determined according to the similarity between the two. If the similarity between the two is large, the adjustment amount of the shared scene weight of the i-th roadside camera is close to the adjusted weight of the roadside camera whose weight has changed. If the similarity between the two is small, the difference between the adjustment amount of the shared scene weight of the i-th roadside camera and the adjusted weight of the roadside camera whose weight has changed is also larger.

[0044] (3) Use the sum of the original shared scene weight of the i-th roadside camera and the change amount as the current shared scene weight of the i-th roadside camera.

[0045] Specifically, an initial shared scene weight can be set for each camera. Among them, the shared scene weight after the change of the initial shared scene weight over time is used as the original shared scene weight set for each camera. For the same camera cluster, as long as the weight of one camera is adjusted, the original shared scene weights of other cameras can change accordingly, thereby reducing the workload of maintaining the shared scene weights of the cameras.

[0046] In this embodiment, through the daily operation and maintenance work of autonomous driving supervision, the selection weights of roadside cameras are dynamically adjusted and calibrated, improving the accuracy of roadside camera selection and avoiding the difficulty of logical adjustment caused by the hard-coding method. At the same time, the requirement for the integrity of camera information is reduced to adapt to complex camera information data. When the camera information is insufficient to accurately obtain the logical coding method, the result of algorithm selection can be effectively intervened through the weight configuration method. The weight data structure with completely independent weights for each camera can effectively support the refined adjustment of key areas and complex road ends.

[0047] 105. Select the videos of a preset number of roadside cameras in the order of decreasing priority.

[0048] In a specific implementation process, the videos of a preset number of roadside cameras can be selected in the order of decreasing priority, so that the required videos can be obtained quickly and accurately, improving the monitoring accuracy and response efficiency in a complex road network environment.

[0049] The method for selecting road camera videos based on dynamic camera weights in this embodiment constructs a rectangular area centered on the longitude and latitude of the vehicle, and divides the rectangular area into 4 sub-areas; uses the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment of the required moment to determine the driving scenario of the vehicle; adjusts the area weights of at least one sub-area according to the driving scenario of the vehicle; determines the priority of each roadside camera by using the driving scenario weight of each roadside camera, the area weight of the sub-area where each roadside camera is located, and the current shared scenario weight of each roadside camera; and selects the videos of a preset number of roadside cameras in the order of decreasing priority. In this way, the camera acquisition algorithm can be more in line with the video selection tendency under different road conditions and different vehicle operating states, avoiding the problem that the fixed algorithm cannot flexibly control the selection expectation, and improving the monitoring accuracy and response efficiency in a complex road network environment.

[0050] Based on the same general inventive concept, the present invention also protects a system for selecting road camera videos based on dynamic camera weights. The system for selecting road camera videos based on dynamic camera weights provided by the present invention will be described below. The system for selecting road camera videos based on dynamic camera weights described below can be mutually referred to the method for selecting road camera videos based on dynamic camera weights described above.

[0051] Figure 6 It is a schematic structural diagram of the system for selecting road camera videos based on dynamic camera weights provided by the embodiments of the present invention, as Figure 6As shown in the figure, the road camera video selection system based on the dynamic weight of the camera in this embodiment includes a construction module 61, a first determination module 62, an adjustment module 63, a second determination module 64, and a selection module 65.

[0052] Among them, the construction module 61 is used to construct a rectangular area centered on the longitude and latitude of the vehicle, and divide the rectangular area into 4 sub-areas; among them, the long axis direction of the rectangular area is consistent with the vehicle driving direction, and the longitude and latitude of multiple roadside cameras are included in the rectangular area; The first determination module 62 is used to determine the driving scenario of the vehicle by using the difference between the first heading angle of the vehicle at the required moment and the second heading angle of the vehicle at the previous moment of the required moment; The adjustment module 63 is used to adjust the area weight of each sub-area according to the driving scenario of the vehicle; The second determination module 64 is used to determine the priority of each roadside camera by using the driving scenario weight of each roadside camera, the area weight of the sub-area where each roadside camera is located, and the current shared scenario weight of each roadside camera; The selection module 65 is used to select the videos of a preset number of roadside cameras in the order of decreasing priority.

[0053] Figure 7 It is a schematic structural diagram of an electronic device provided by an embodiment of the present invention. The road camera video selection system based on the dynamic weight of the camera may include: a processor 710, a communication interface 720, a memory 730, and a communication bus 740. Among them, the processor 710, the communication interface 720, and the memory 730 complete mutual communication through the communication bus 740. The processor 710 can call the logical instructions in the memory 730 to execute the road camera video selection method based on the dynamic weight of the camera.

[0054] In addition, when the logical instructions in the above-mentioned memory 730 are implemented in the form of software functional units and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which may be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods described in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memories (ROM, Read-Only Memory), random access memories (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0055] On the other hand, the present invention also provides a computer program product. The computer program product includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute the method for selecting road camera videos based on dynamic weights of a camera provided by the above-mentioned various methods.

[0056] On another aspect, the present invention also provides a non-transitory computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it is implemented to execute the method for selecting road camera videos based on dynamic weights of a camera provided by the above-mentioned various methods.

[0057] It should be noted that the relevant user personal information that may be involved in various embodiments of this application is all processed in strict accordance with the requirements of laws and regulations, following the principles of legality, legitimacy, and necessity, for reasonable purposes based on business scenarios, and is personal information actively provided by users during the process of using products / services or generated due to the use of products / services, as well as personal information obtained with the authorization of users.

[0058] The user personal information processed by this application will vary depending on the specific product / service scenario. It is subject to the specific scenario of the user using the product / service and may involve the user's account information, device information, driving information, vehicle information, or other relevant information. This application will treat the user's personal information and its processing with a high degree of diligence.

[0059] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed to multiple network units. Some or all of the modules can be selected according to actual needs to achieve the purpose of the solution of this embodiment. A person of ordinary skill in the art can understand and implement it without creative work.

[0060] Through the description of the above embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus a necessary general hardware platform, and of course, it can also be implemented by hardware. Based on this understanding, the essence of the above technical solution, or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute the methods described in each embodiment or some parts of the embodiments.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements for some of the technical features. And these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for selecting road camera videos based on dynamic weights of cameras, characterized in that, Including: Construct a rectangular area centered on the longitude and latitude of the vehicle, and divide the rectangular area into 4 sub-areas; wherein, the long axis direction of the rectangular area is consistent with the vehicle driving direction, and the mapped longitude and latitude of multiple roadside cameras are included in the rectangular area; Use the difference between the first heading angle of the vehicle at the demand moment and the second heading angle of the vehicle at the previous moment of the demand moment to determine the driving scenario of the vehicle; Adjust the area weights of at least one sub-area according to the driving scenario of the vehicle; Use the driving scenario weight of each roadside camera, the area weight of the sub-area where each roadside camera is located, and the current shared scenario weight of each roadside camera to determine the priority of each roadside camera; Select the videos of a preset number of roadside cameras in the order from high to low priority.

2. The method for selecting road camera videos based on dynamic weights of a camera according to claim 1, wherein Using the difference between the first heading angle of the vehicle at the demand moment and the second heading angle of the vehicle at the previous moment of the demand moment to determine the driving scenario of the vehicle, including: If the absolute value of the difference is less than or equal to a preset absolute value, determine the driving scenario of the vehicle according to the scenario range value to which the difference belongs; If the absolute value of the difference is greater than the preset absolute value, add or subtract the difference from a preset value to obtain a conversion value corresponding to the difference, and determine the driving scenario of the vehicle according to the scenario range value to which the conversion value belongs.

3. The method for selecting road camera videos based on dynamic camera weights according to claim 2, wherein If the target value is greater than the first preset difference of the scenario range value and less than the second preset difference of the scenario range value, determine that the driving scenario of the vehicle is a right turn; wherein, the first preset difference is a positive number; wherein, the target value includes the difference or the conversion value; If the target value is greater than or equal to the third preset difference of the scenario range value and less than or equal to the first preset difference, determine that the driving scenario of the vehicle is a straight line; wherein, the third preset difference is a negative number; If the target value is greater than the fourth preset difference of the scenario range value and less than the third preset difference, determine that the driving scenario of the vehicle is a left turn; wherein, the preset absolute value is the difference between the second preset difference and the fourth preset difference.

4. The method for selecting road camera videos based on dynamic weights of cameras according to claim 1, characterized in that, Adjusting the area weights of at least one sub-area according to the driving scenario of the vehicle, including: According to the driving scenario of the vehicle, use the coincidence degree between the shooting range of each roadside camera and the vehicle trajectory to predict the predicted shooting duration of the vehicle in each sub-area; Select the at least one sub-area in the order from high to low predicted shooting duration, and adjust the area weights of the at least one sub-area.

5. The method for selecting road camera videos based on dynamic weights of cameras according to claim 4, wherein, It also includes: If the driving scenario of the vehicle is a left turn or a right turn, expand the width of the at least one sub-area by a preset multiple, or If the driving scenario of the vehicle is a left turn or a right turn, obtain the turning radius of the vehicle, determine the expansion coefficient based on the preset correlation between the turning radius and the expansion coefficient, and expand the width of the at least one sub-region according to the expansion coefficient.

6. The method for selecting road camera videos based on dynamic weights of a camera according to claim 1, wherein Determine the priority of each roadside camera by using the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the shared scenario weight of each roadside camera, including: Calculate the product of the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the shared scenario weight of each roadside camera; Set the priority of each roadside camera in descending order of the product.

7. The method for selecting road camera videos based on dynamic weights of cameras according to any one of claims 1-6, characterized in that, It also includes: If the weight of any target camera among the remaining cameras in the camera cluster to which the i-th roadside camera in the multiple roadside cameras belongs changes, obtain the adjusted weight of the any target camera; wherein, the camera cluster is obtained by clustering all the roadside cameras to be processed; Use the adjusted weight to determine the change amount of the shared scenario weight of the i-th roadside camera; Take the sum of the original shared scenario weight of the i-th roadside camera and the change amount as the current shared scenario weight of the i-th roadside camera.

8. The method for selecting road camera videos based on dynamic weights of cameras according to claim 7, wherein Using the adjusted weight to determine the change amount of the shared scenario weight of the i-th roadside camera includes: Determine the similarity between the i-th roadside camera and the target camera; Determine the weight transfer coefficient corresponding to the i-th roadside camera according to the preset correlation between the similarity and the weight transfer coefficient; Take the product of the adjusted weight and the weight transfer coefficient as the change amount.

9. A road camera video selection system based on dynamic weights of cameras, characterized in that, It includes: A construction module for constructing a rectangular area centered on the longitude and latitude of the vehicle and dividing the rectangular area into 4 sub-regions; wherein, the long axis direction of the rectangular area is consistent with the vehicle driving direction, and the longitude and latitude of multiple roadside cameras are included in the rectangular area; A first determination module for determining the driving scenario of the vehicle by using the difference between the first heading angle of the vehicle at the demand moment and the second heading angle of the vehicle at the previous moment of the demand moment; An adjustment module for adjusting the regional weight of each sub-region according to the driving scenario of the vehicle; A second determination module for determining the priority of each roadside camera by using the driving scenario weight of each roadside camera, the regional weight of the sub-region where each roadside camera is located, and the current shared scenario weight of each roadside camera; A selection module for selecting the videos of a preset number of roadside cameras in descending order of priority.

10. An electronic device, characterized in that, It includes a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the method for selecting road camera videos based on dynamic camera weights as described in any one of claims 1-8.

Images