A road monitoring video selection method based on vehicle position and heading

Abstract

The present invention provides a road monitoring video selection method based on vehicle position and heading, which relates to the field of intelligent transportation technology, including the following steps: analyzing the shooting area of ​​the current target camera to obtain a target shooting area; predicting the estimated position of the target vehicle at a future moment according to the real-time running status data of the target vehicle detected in real time; combining the real-time position and the estimated position in the real-time status data with the target shooting area for analysis, and performing target camera screening and corresponding video pulling for the target vehicle at the current moment and future moment. By obtaining the target shooting area of ​​the target camera; combining the real-time position and the estimated position of the target vehicle at a future moment predicted according to the real-time running status data of the target vehicle with the target shooting area for analysis, and performing target camera screening and corresponding video pulling for the target vehicle at the current moment and future moment, the accuracy of real-time monitoring and the effectiveness of video acquisition can be effectively improved.

Description

Technical Field

[0001] The present invention relates to the field of intelligent transportation technology, and in particular to a road monitoring video selection method based on vehicle position and heading. Background Art

[0002] Traditional road monitoring systems usually rely on fixed cameras for monitoring. However, in complex traffic environments, the monitoring range of a single camera is limited, making it difficult to fully track the target vehicle. Existing cross-camera vehicle tracking methods mostly rely on manual operation or real-time vehicle GPS data, which are cumbersome to operate and have low applicability. Therefore, how to accurately pull road camera videos based on the camera's shooting range and vehicle position to achieve accurate vehicle video monitoring has become one of the current research focuses.

[0003] Therefore, the present invention provides a road monitoring video selection method based on vehicle position and heading. Summary of the invention

[0004] The present invention provides a road monitoring video selection method based on vehicle position and heading, which is used to obtain a target shooting area of ​​a target camera; the real-time position and the estimated position of the target vehicle at a future moment predicted according to the real-time running status data of the target vehicle are combined with the target shooting area for analysis to perform target camera screening and corresponding video pulling at the current moment and future moment of the target vehicle, which can effectively improve the accuracy of real-time monitoring and the effectiveness of video acquisition.

[0005] The present invention provides a road monitoring video selection method based on vehicle position and heading, comprising:

[0006] Step 1: Obtain basic information of the target camera and process it to obtain the first key data;

[0007] Step 2: According to the first key data, the shooting area of ​​the current target camera is analyzed to obtain the target shooting area;

[0008] Step 3: Detect the operating status of the target vehicle in real time, and predict the estimated position of the target vehicle at a future time based on the acquired real-time operating status data;

[0009] Step 4: Combine the real-time position and estimated position in the real-time status data with the target shooting area for analysis, and perform target camera screening and corresponding video extraction for the target vehicle at the current moment and in the future.

[0010] Preferably, basic information of the target camera is obtained and processed to obtain first key data, including:

[0011] Extracting a position content description from the basic information of the current target camera, and performing an orientation description on the acquired position content description according to a set standard description rule to obtain a first orientation position;

[0012] Combine the camera number in the basic information with the first position, and name the head pole of the current target camera according to the set head pole naming rule;

[0013] The first orientation position of the target camera, the camera pole name and basic information are output as the first key data.

[0014] Preferably, according to the first key data, a shooting area analysis is performed on the current target camera to obtain the target shooting area, including:

[0015] Using a preset measurement tool, measure the relative position of the head pole of the current target camera and the preset reference point to obtain first relative position information;

[0016] Inputting the first relative position information and the first azimuth position of the target camera into a pre-established angle analysis model to obtain a first possible shooting direction angle of the current target camera;

[0017] Using a set angle measuring device to measure the second possible shooting direction angle of the current target camera pointing to the opposite pole;

[0018] Combining and analyzing the first possible shooting direction angle and the second possible shooting direction angle, to obtain a target shooting direction angle;

[0019] The target shooting direction angle is combined with the fixed shooting angle range of the current target camera to calculate and obtain the target shooting range;

[0020] Determine a target shooting area of ​​the current target camera based on the target shooting range;

[0021] The target shooting direction angle, target shooting range, shooting direction and target shooting area set are used as key shooting data and filled into the corresponding specific position of the camera database to which the current target camera belongs.

[0022] Preferably, combining and analyzing the first possible shooting direction angle and the second possible shooting direction angle to obtain the target shooting direction angle includes:

[0023] Compare the first possible shooting direction angle with the second possible shooting direction angle to obtain the absolute difference of the direction angles;

[0024] When the direction angle absolute difference is less than the set angle difference threshold lower limit, calculating the average value of the first possible shooting direction angle and the second possible shooting direction angle, and outputting it as the target shooting direction angle;

[0025] When the direction angle absolute difference falls within the set angle difference threshold range, performing a collective adjustment calculation on the first possible shooting direction angle and the second possible shooting direction angle to obtain an adjusted direction angle, and outputting the adjusted direction angle as a target shooting direction angle;

[0026] The calculation formula of the adjusted direction angle is as follows:

[0027] ; In the formula, It is expressed as the adjusted direction angle; It is expressed as the first possible shooting direction angle; It is expressed as the second possible shooting direction angle; It is expressed as the measurement error rate of the set angle measurement device; It is expressed as the model test failure rate of the angle analysis model; It is expressed as natural logarithm; e is expressed as a constant with a value of 2.7;

[0028] When the absolute difference in the direction angle is greater than the set angle difference threshold upper limit, the first possible shooting direction angle and the second possible shooting direction angle are recalculated and combined for analysis until the target shooting direction angle is obtained.

[0029] Preferably, real-time detection of the operating status of the target vehicle and prediction of the estimated position of the target vehicle at a future time based on the acquired real-time operating status data include:

[0030] Use preset sensors to detect the operating status of the target vehicle in real time and obtain real-time operating status data;

[0031] After pre-processing the operating status data, the operating status data is converted into a unified data format to obtain first status data;

[0032] Extracting condition data related to the predicted vehicle position from the first condition data and outputting the data as real-time associated data;

[0033] The real-time associated data is input into a pre-established position prediction model to predict the vehicle position of the current target vehicle at a future time to obtain an estimated position.

[0034] Preferably, the real-time associated data refers to the real-time position, real-time speed and real-time heading angle of the target vehicle.

[0035] Preferably, the real-time position and estimated position in the real-time status data are combined with the target shooting area for analysis, and target camera screening and corresponding video pulling are performed for the current moment and future moment of the target vehicle, including:

[0036] According to the real-time position of the target vehicle, query the camera database within the set range to obtain the first camera database;

[0037] Mark all target cameras in the first camera database as cameras to be selected;

[0038] According to the first position of the camera to be selected, all the cameras to be selected are sorted from largest to smallest according to the distance from the current target vehicle to obtain a camera list;

[0039] Traversing each candidate camera in the camera list in turn, marking the candidate camera whose real-time position of the target vehicle is in the corresponding target shooting area as the first camera;

[0040] Calculate the first angle between the real-time heading angle of the current target vehicle and the target shooting direction angle of each candidate camera in the camera list;

[0041] The candidate camera whose first included angle between the target shooting direction angle and the real-time heading angle of the current target vehicle falls within the set included angle threshold range is marked as the second camera;

[0042] Re-mark the candidate camera marked as both the first camera and the second camera as the priority camera;

[0043] Performing priority analysis and corresponding sorting on the first camera, the second camera and the priority camera to obtain a priority camera list;

[0044] Filter the first two cameras from the priority camera list as applicable cameras;

[0045] Sending a video pull request to the applicable camera to obtain real-time road monitoring video;

[0046] The captured road surveillance video is displayed in real time on the target surveillance screen, and stored and backed up as needed;

[0047] According to the estimated position of the target vehicle, query the camera database within the set range to obtain the second camera database;

[0048] Mark all target cameras in the second camera database as pre-selected cameras;

[0049] According to the first position of the pre-selected camera, all pre-selected cameras are sorted from largest to smallest according to the estimated position of the current target vehicle to obtain a pre-selected camera list;

[0050] Traversing each pre-selected camera in the pre-selected camera list in turn, deleting the pre-selected camera whose estimated position of the target vehicle is not in the corresponding target shooting area from the pre-selected camera list, and obtaining a new pre-selected camera list;

[0051] The first three pre-selected cameras are selected from the new pre-selected camera list as the reserved cameras, and the real-time video streams of the reserved cameras are pulled in advance.

[0052] Preferably, the first camera, the second camera and the first-priority camera are analyzed and sorted accordingly to obtain a priority camera list, including:

[0053] Calculating a corresponding priority coefficient for the priority camera based on the first angle and the distance from the current target vehicle;

[0054] The calculation formula of the priority coefficient is as follows:

[0055] ; In the formula, It is expressed as the priority coefficient of the current priority camera; It is represented as the first angle between the target shooting direction angle of the current priority camera and the real-time heading angle of the current target vehicle; It represents the upper limit of the angle threshold range; It represents the lower limit of the angle threshold range; It is expressed as the influence weight of the angle between the shooting angle and the real-time heading angle of the vehicle on the priority of the analysis camera; It is represented by the real-time distance between the current priority camera and the target vehicle; It is expressed as the sum of the real-time distances between all priority cameras and the target vehicle; It is expressed as the weight of the distance between the camera and the target vehicle on the priority of the analysis camera;

[0056] Sort all priority cameras from large to small according to their priority coefficients to get the first priority column;

[0057] Sort all first cameras in ascending order of distance from the current target vehicle to obtain a second priority column;

[0058] All second cameras are sorted from small to large according to the absolute value of the first angle between the corresponding target shooting direction angle and the real-time heading angle of the current target vehicle to obtain a third priority column;

[0059] The first priority list, the second priority list and the third priority list are sequentially concatenated to obtain a priority camera list.

[0060] Compared with the prior art, the present invention has the following beneficial effects:

[0061] By obtaining the target shooting area of ​​the target camera; combining the real-time position and the estimated position of the target vehicle in the future predicted based on the real-time operating status data of the target vehicle with the target shooting area for analysis, target camera screening and corresponding video pulling are performed on the target vehicle at the current moment and future moment, which can effectively improve the accuracy of real-time monitoring and the effectiveness of video acquisition.

[0062] Other features and advantages of the present invention will be described in the following description, and partly become apparent from the description, or understood by practicing the present invention. The purpose and other advantages of the present invention can be realized and obtained by the structures particularly pointed out in the written description and the accompanying drawings.

[0063] The technical solution of the present invention is further described in detail below through the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS

[0064] The accompanying drawings are used to provide a further understanding of the present invention and constitute a part of the specification. Together with the embodiments of the present invention, they are used to explain the present invention and do not constitute a limitation of the present invention. In the accompanying drawings:

[0065] Figure 1 The present invention is a flowchart of a method for selecting road monitoring videos based on vehicle position and heading in an embodiment of the present invention. DETAILED DESCRIPTION

[0066] The preferred embodiments of the present invention are described below in conjunction with the accompanying drawings. It should be understood that the preferred embodiments described herein are only used to illustrate and explain the present invention, and are not used to limit the present invention.

[0067] The embodiment of the present invention provides a method for selecting road monitoring videos based on vehicle position and heading, such as Figure 1 As shown, including:

[0068] Step 1: Obtain basic information of the target camera and process it to obtain the first key data;

[0069] Step 2: According to the first key data, the shooting area of ​​the current target camera is analyzed to obtain the target shooting area;

[0070] Step 3: Detect the operating status of the target vehicle in real time, and predict the estimated position of the target vehicle at a future time based on the acquired real-time operating status data;

[0071] Step 4: Combine the real-time position and estimated position in the real-time status data with the target shooting area for analysis, and perform target camera screening and corresponding video extraction for the target vehicle at the current moment and in the future.

[0072] In this embodiment, the basic information includes camera number, location content description, model, resolution, preset reference point, opposite pole and installation conditions (such as ground inclination), etc.; the target camera refers to a specific camera used to monitor the road and requires video pulling and selection; the first key data refers to the first azimuth position of the target camera, the camera pole name and basic information; the target shooting area refers to the area that the target camera can actually shoot; the real-time operating status data refers to the real-time operating data of the target vehicle at the current moment, such as the current position, speed, acceleration, heading, etc.; the estimated position refers to the predicted position of the target vehicle in the future, such as the position in the next 1 second; the real-time position refers to the position of the target vehicle at the current moment.

[0073] The beneficial effects of the above technical solution are: by acquiring the target shooting area of ​​the target camera; combining the real-time position and the estimated position of the target vehicle at a future moment predicted based on the real-time operating status data of the target vehicle with the target shooting area for analysis, target camera screening and corresponding video pulling are performed on the target vehicle at the current moment and future moment, which can effectively improve the accuracy of real-time monitoring and the effectiveness of video acquisition.

[0074] The embodiment of the present invention provides a road monitoring video selection method based on vehicle position and heading, which obtains and processes basic information of a target camera to obtain first key data, including:

[0075] Extracting a position content description from the basic information of the current target camera, and performing an orientation description on the acquired position content description according to a set standard description rule to obtain a first orientation position;

[0076] Combine the camera number in the basic information with the first position, and name the head pole of the current target camera according to the set head pole naming rule;

[0077] The first orientation position of the target camera, the camera pole name and basic information are output as the first key data.

[0078] In this embodiment, the location content description refers to a detailed description of the installation location of the target camera, including longitude and latitude, address, etc.; setting a standard description rule refers to a rule used to standardize the location content description, and describes the location of the target camera in an azimuth manner (such as east, west, south, and north) according to the acquired longitude and latitude; the first azimuth position refers to the camera location description processed according to the setting standard description rule; setting a head pole naming rule refers to a rule used to name the camera pole, which is based on the location and numbering setting of the camera, generally a location prefix + a numbering part, such as the head pole East District 001 of the target camera, which is represented by a camera pole located in the East District and numbered 001; the head pole of the target camera refers to the pole on which the target camera is installed; the camera pole naming refers to the naming information of the head pole of the current target camera.

[0079] The beneficial effect of the above technical solution is: by processing and analyzing the basic information of the acquired target camera to obtain the first key data, a strong data basis can be provided for the subsequent determination of the target shooting area of ​​the target camera.

[0080] The embodiment of the present invention provides a method for selecting a road monitoring video based on a vehicle position and heading, and according to the first key data, a shooting area analysis is performed on the current target camera to obtain the target shooting area, including:

[0081] Using a preset measurement tool, measure the relative position of the head pole of the current target camera and the preset reference point to obtain first relative position information;

[0082] Inputting the first relative position information and the first azimuth position of the target camera into a pre-established angle analysis model to obtain a first possible shooting direction angle of the current target camera;

[0083] Using a set angle measuring device to measure the second possible shooting direction angle of the current target camera pointing to the opposite pole;

[0084] Combining and analyzing the first possible shooting direction angle and the second possible shooting direction angle, to obtain a target shooting direction angle;

[0085] The target shooting direction angle is combined with the fixed shooting angle range of the current target camera to calculate and obtain the target shooting range;

[0086] Determine a target shooting area of ​​the current target camera based on the target shooting range;

[0087] The target shooting direction angle, target shooting range, shooting direction and target shooting area set are used as key shooting data and filled into the corresponding specific position of the camera database to which the current target camera belongs.

[0088] In this embodiment, the preset measuring tool refers to a tool used to measure the relative position between the head pole of the target camera and a preset reference point, such as a laser rangefinder or a GPS locator; the preset reference point refers to a predetermined fixed reference object used to determine the position of the target camera; the first relative position information refers to the relative position relationship between the head pole of the target camera and the preset reference point measured by the preset measuring tool, including information such as distance and direction angle.

[0089] In this embodiment, the angle analysis model refers to a model obtained by pre-training a neural network, which calculates the shooting direction angle of the target camera based on input information (such as first relative position information, first azimuth position). It is obtained by collecting a large amount of data about the camera position, direction and corresponding shooting direction angle, as well as parameters such as the camera's installation position, height, and tilt angle, and then extracting useful features from the collected data (such as the camera's relative position information (such as distance, direction angle), azimuth position, installation height, and tilt angle) and combining the collected data as training data to train the neural network; the first possible shooting direction angle refers to the possible shooting direction angle of the current target camera calculated by the angle analysis model.

[0090] In this embodiment, the set angle measurement device refers to a device used to measure the angle of the target camera pointing to the opposite pole, generally refers to an angle measuring instrument, wherein the opposite pole refers to a reference pole predetermined within the camera field of view for assisting in measuring the camera shooting direction; the second possible shooting direction angle refers to the angle of the current target camera pointing to the opposite pole measured by the set angle measurement device; the fixed shooting angle range refers to the shooting angle range that a fixed camera can capture, for example, a 60 degree left and right offset; the target shooting range refers to the range that the camera can actually capture calculated by combining the target shooting direction angle and the fixed shooting angle range; the target shooting area refers to the target shooting range, the actual installation position and height of the target camera, and the ground inclination as input, and the input of the preset The area generation model established first generates a specific area that can be clearly captured by the camera. Among them, the ground inclination is obtained from the installation environment conditions in the basic information of the target camera. The area generation model needs to collect a large amount of data about the camera installation environment (including the camera's physical parameters (such as focal length, field of view), installation location (latitude and longitude, altitude), shooting direction, ground inclination, etc.) and the actual captured image data, and all the collected data are preprocessed and key features are extracted. The extracted key features (such as installation height, shooting direction, ground inclination) are combined into feature vectors as input for model training to train the neural network; the key shooting data is composed of the target shooting direction angle, target shooting range, shooting direction and target shooting area set.

[0091] In this embodiment, for example, there is a fixed shooting angle range of the target camera 1, which is 60 degrees to the left and right (that is, the total offset angle is 120 degrees, from -60 degrees to +60 degrees, relative to the reference direction 1), and the target shooting direction angle is 70° (relative to the reference direction 1).

[0092] At this time, the left offset angle of the target shooting range is 10° ( , relative to the reference direction 1), the right offset angle is 130° ( , relative to the reference direction 1), that is, the target shooting range is (relative to reference direction 1).

[0093] The beneficial effect of the above technical solution is: by analyzing the shooting area of ​​the current target camera according to the acquired first key data to obtain the target shooting area, the monitoring accuracy of the camera can be improved, and data support can be provided for the screening of subsequent cameras.

[0094] The embodiment of the present invention provides a road monitoring video selection method based on vehicle position and heading, which combines and analyzes a first possible shooting direction angle and a second possible shooting direction angle to obtain a target shooting direction angle, including:

[0095] Compare the first possible shooting direction angle with the second possible shooting direction angle to obtain the absolute difference of the direction angles;

[0096] When the direction angle absolute difference is less than the set angle difference threshold lower limit, calculating the average value of the first possible shooting direction angle and the second possible shooting direction angle, and outputting it as the target shooting direction angle;

[0097] When the direction angle absolute difference falls within the set angle difference threshold range, performing a collective adjustment calculation on the first possible shooting direction angle and the second possible shooting direction angle to obtain an adjusted direction angle, and outputting the adjusted direction angle as a target shooting direction angle;

[0098] The calculation formula of the adjusted direction angle is as follows:

[0099] ; In the formula, It is expressed as the adjusted direction angle; It is expressed as the first possible shooting direction angle; It is expressed as the second possible shooting direction angle; It is expressed as the measurement error rate of the set angle measurement device; It is expressed as the model test failure rate of the angle analysis model; It is expressed as natural logarithm; e is expressed as a constant with a value of 2.7;

[0100] When the absolute difference in the direction angle is greater than the set angle difference threshold upper limit, the first possible shooting direction angle and the second possible shooting direction angle are recalculated and combined for analysis until the target shooting direction angle is obtained.

[0101] In this embodiment, the direction angle absolute difference refers to the absolute value of the angle difference between the first possible shooting direction angle and the second possible shooting direction angle; the angle difference threshold range is composed of the angle difference threshold lower limit and the angle difference threshold upper limit, which generally refers to ; The measurement error rate of the angle measurement device is set on the basis of the set allowable measurement error range, and is the ratio of measurement errors within a preset time period obtained in advance through experiments and statistical analysis, that is, the number of test errors / total number of tests, wherein the preset time period is preset, such as 3 months; The model test error rate of the angle analysis model is based on the set allowable measurement error range, and is obtained by testing the model with a preset data set, that is, the number of test errors / total number of tests, wherein the data set consists of the azimuth positions of several different cameras, and the relative positions of the head pole and the corresponding preset reference points; Setting the allowable measurement error range refers to the acceptable measurement error range, that is, the error of the angle measurement is within the set allowable measurement error range, and the current angle measurement is judged to be correct, generally The adjusted direction angle is an angle calculated based on the first possible shooting direction angle, the second possible shooting direction angle, the measurement error rate of the set angle measurement device, and the model test error rate of the angle analysis model.

[0102] The beneficial effect of the above technical solution is: by comprehensively considering different possible shooting direction angles and related error rates, a more accurate and reliable target shooting direction angle is obtained, which can help to accurately obtain the target shooting area of ​​the current target camera.

[0103] The embodiment of the present invention provides a road monitoring video selection method based on vehicle position and heading, which detects the running status of a target vehicle in real time, and predicts the estimated position of the target vehicle at a future time according to the acquired real-time running status data, including:

[0104] Use preset sensors to detect the operating status of the target vehicle in real time and obtain real-time operating status data;

[0105] After pre-processing the operating status data, the operating status data is converted into a unified data format to obtain first status data;

[0106] Extracting condition data related to the predicted vehicle position from the first condition data and outputting the data as real-time associated data;

[0107] The real-time associated data is input into a pre-established position prediction model to predict the vehicle position of the current target vehicle at a future time to obtain an estimated position.

[0108] In this embodiment, the preset sensor refers to a sensor device installed on the vehicle, such as a speed sensor, which is used to detect the operating status of the target vehicle in real time; the real-time operating status data refers to the operating status information of the target vehicle captured and transmitted in real time by the preset sensor, including the vehicle's current position, speed (instantaneous speed, average speed, etc.), acceleration, heading angle (the angle between the vehicle's driving direction and the geographic direction), etc.

[0109] In this embodiment, data preprocessing refers to operations such as cleaning, denoising, calibration and formatting of real-time operating status data; unified data format refers to converting the preprocessed real-time operating status data into a standard data format to facilitate subsequent data processing and analysis; first status data refers to real-time operating status data after data preprocessing and conversion into a unified data format; real-time associated data refers to the real-time position, real-time speed and real-time heading angle of the target vehicle; the position prediction model refers to a pre-established model for predicting the position of the target vehicle at a future time based on the real-time associated data, which is obtained by collecting a large amount of vehicle operating status data (including position, speed, acceleration, steering wheel angle, wheel speed, etc.), preprocessing the collected data, extracting features related to the predicted vehicle position (such as historical position, speed change, acceleration change, steering wheel angle change, etc.), and then using the extracted features and the processed data as training data to train the neural network; the estimated position refers to the predicted position of the target vehicle at a future time calculated by the position prediction model based on the real-time associated data, such as the position in the next 1 second.

[0110] The beneficial effect of the above technical solution is: by obtaining real-time operating status data based on real-time detection of the operating status of the target vehicle, the estimated position of the target vehicle at a future moment is predicted, which can help optimize the allocation of monitoring resources and improve monitoring efficiency and performance.

[0111] The embodiment of the present invention provides a road monitoring video selection method based on vehicle position and heading, which combines the real-time position and estimated position in the real-time status data with the target shooting area for analysis, and performs target camera screening and corresponding video extraction at the current moment and future moment of the target vehicle, including:

[0112] According to the real-time position of the target vehicle, query the camera database within the set range to obtain the first camera database;

[0113] Mark all target cameras in the first camera database as cameras to be selected;

[0114] According to the first position of the camera to be selected, all the cameras to be selected are sorted from largest to smallest according to the distance from the current target vehicle to obtain a camera list;

[0115] Traversing each candidate camera in the camera list in turn, marking the candidate camera whose real-time position of the target vehicle is in the corresponding target shooting area as the first camera;

[0116] Calculate the first angle between the real-time heading angle of the current target vehicle and the target shooting direction angle of each candidate camera in the camera list;

[0117] The camera to be selected, whose first angle between the target shooting direction angle and the real-time heading angle of the current target vehicle falls within the set angle threshold range, is marked as the second camera;

[0118] Re-mark the candidate camera marked as both the first camera and the second camera as the priority camera;

[0119] Performing priority analysis and corresponding sorting on the first camera, the second camera and the priority camera to obtain a priority camera list;

[0120] Filter the first two cameras from the priority camera list as applicable cameras;

[0121] Sending a video pull request to the applicable camera to obtain real-time road monitoring video;

[0122] The captured road surveillance video is displayed in real time on the target surveillance screen, and stored and backed up as needed;

[0123] According to the estimated position of the target vehicle, query the camera database within the set range to obtain the second camera database;

[0124] Mark all target cameras in the second camera database as pre-selected cameras;

[0125] According to the first position of the pre-selected camera, all pre-selected cameras are sorted from largest to smallest according to the estimated position of the current target vehicle to obtain a pre-selected camera list;

[0126] Traversing each pre-selected camera in the pre-selected camera list in turn, deleting the pre-selected camera whose estimated position of the target vehicle is not in the corresponding target shooting area from the pre-selected camera list, and obtaining a new pre-selected camera list;

[0127] The first three pre-selected cameras are selected from the new pre-selected camera list as the reserved cameras, and the real-time video streams of the reserved cameras are pulled in advance.

[0128] In this embodiment, the set range refers to a semicircular range with the vehicle position as the starting point and the set distance as the radius, wherein the set distance is predetermined; the first camera database refers to a camera database within the set range determined according to the real-time position of the target vehicle, wherein the camera database is used to store basic information and key shooting data of different target cameras (such as the target shooting direction angle, the first shooting direction and the target shooting area); the camera to be selected refers to the camera included in the first camera database; the camera list refers to a list obtained by sorting the cameras to be selected according to their distance from the target vehicle.

[0129] In this embodiment, the first camera refers to a camera to be selected whose real-time position of the target vehicle is in the corresponding target shooting area; the first angle refers to the angle between the real-time heading angle of the target vehicle and the target shooting direction angle of the camera to be selected; the angle threshold range is predetermined, for example, ; The priority camera refers to the candidate camera that satisfies the requirements that the real-time position of the target vehicle is within the corresponding shooting area and the first angle between the shooting angle and the heading angle of the target vehicle is within the set angle threshold range; the priority camera list is obtained by sorting the first camera, the second camera and the priority camera after priority analysis; the applicable camera refers to the first two cameras selected from the priority camera list; the target monitoring screen refers to the screen used to display the pulled road monitoring video.

[0130] In this embodiment, the second camera database refers to a camera database within a set range determined according to the estimated position of the target vehicle; the pre-selected cameras refer to the cameras included in the second camera database; the pre-selected camera list refers to a list obtained by sorting the pre-selected cameras according to their distance from the target vehicle; the new pre-selected camera list is obtained by deleting the pre-selected cameras whose estimated positions of the target vehicle are not in the corresponding target shooting area from the pre-selected camera list; the predetermined cameras refer to the first three pre-selected cameras screened out from the new pre-selected camera list.

[0131] The beneficial effect of the above technical solution is: by combining the real-time position and estimated position in the real-time status data with the target shooting area for analysis, target camera screening and corresponding video pulling are performed for the target vehicle at the current moment and future moment, which can effectively improve the accuracy of real-time monitoring and the effectiveness of video acquisition.

[0132] The embodiment of the present invention provides a method for selecting road surveillance videos based on vehicle position and heading, and performs priority analysis and corresponding sorting on the first camera, the second camera and the first priority camera to obtain a priority camera list, including:

[0133] Calculating a corresponding priority coefficient for the priority camera based on the first angle and the distance from the current target vehicle;

[0134] The calculation formula of the priority coefficient is as follows:

[0135] ; In the formula, It is expressed as the priority coefficient of the current priority camera; It is represented as the first angle between the target shooting direction angle of the current priority camera and the real-time heading angle of the current target vehicle; It represents the upper limit of the angle threshold range; It represents the lower limit of the angle threshold range; It is expressed as the influence weight of the angle between the shooting angle and the real-time heading angle of the vehicle on the priority of the analysis camera; It is represented by the real-time distance between the current priority camera and the target vehicle; It is expressed as the sum of the real-time distances between all priority cameras and the target vehicle; It is expressed as the weight of the distance between the camera and the target vehicle on the priority of the analysis camera;

[0136] Sort all priority cameras from large to small according to their priority coefficients to get the first priority column;

[0137] Sort all first cameras in ascending order of distance from the current target vehicle to obtain a second priority column;

[0138] All second cameras are sorted from small to large according to the absolute value of the first angle between the corresponding target shooting direction angle and the real-time heading angle of the current target vehicle to obtain a third priority column;

[0139] The first priority list, the second priority list and the third priority list are sequentially concatenated to obtain a priority camera list.

[0140] In this embodiment, the priority coefficient refers to a numerical value used to measure the priority of a camera; the first priority column refers to a list obtained by sorting all priority cameras according to the priority coefficient; the second priority column refers to a list obtained by sorting all first cameras from small to large according to the distance from the current target vehicle; the third priority column refers to a list obtained by sorting all second cameras from small to large according to the absolute value of the first angle between the corresponding target shooting direction angle and the real-time heading angle of the current target vehicle; the priority camera list is a list obtained by splicing the first priority column, the second priority column and the third priority column in sequence.

[0141] In this embodiment, for example, there is a first priority column G1 in which the camera sequence is represented as camera g1, camera g2, and camera g3, a second priority column J1 in which the camera sequence is represented as camera j1, camera j2, and a third priority column K1 in which the camera sequence is represented as camera k1, camera k2, and camera k3;

[0142] At this time, the order of the cameras in the priority camera list is represented as camera g1, camera g2, camera g3, camera j1, camera j2, camera k1, camera k2, and camera k3.

[0143] In this embodiment, the weights assigned to the angle between the shooting angle and the real-time heading angle of the vehicle and the distance between the camera and the target vehicle are obtained by solving a matrix constructed after pairwise comparison and relative importance scoring using the hierarchical analysis method.

[0144] The beneficial effect of the above technical solution is: by performing priority analysis and corresponding sorting on the first camera, the second camera and the priority camera, a priority camera list is obtained, which can optimize resource allocation and improve monitoring efficiency.

[0145] Obviously, those skilled in the art can make various changes and modifications to the present invention without departing from the spirit and scope of the present invention. Thus, if these modifications and variations of the present invention fall within the scope of the claims of the present invention and their equivalents, the present invention is also intended to include these modifications and variations.

Claims

1. A road monitoring video selection method based on vehicle position and heading, characterized in that: include: Step 1: Obtain basic information of the target camera and process it to obtain the first key data; Step 2: According to the first key data, the shooting area of ​​the current target camera is analyzed to obtain the target shooting area; Step 3: Detect the operating status of the target vehicle in real time, and predict the estimated position of the target vehicle at a future time based on the acquired real-time operating status data; Step 4: Combine the real-time position and estimated position in the real-time status data with the target shooting area for analysis, and perform target camera screening and corresponding video extraction for the target vehicle at the current moment and in the future; The real-time position and estimated position in the real-time status data are combined with the target shooting area for analysis, and the target camera is screened and the corresponding video is pulled for the target vehicle at the current moment and in the future, including: According to the real-time position of the target vehicle, query the camera database within the set range to obtain the first camera database; Mark all target cameras in the first camera database as cameras to be selected; According to the first position of the camera to be selected, all the cameras to be selected are sorted from largest to smallest according to the distance from the current target vehicle to obtain a camera list; Traversing each candidate camera in the camera list in turn, marking the candidate camera whose real-time position of the target vehicle is in the corresponding target shooting area as the first camera; Calculate the first angle between the real-time heading angle of the current target vehicle and the target shooting direction angle of each candidate camera in the camera list; The candidate camera whose first included angle between the target shooting direction angle and the real-time heading angle of the current target vehicle falls within the set included angle threshold range is marked as the second camera; Re-mark the candidate camera marked as both the first camera and the second camera as the priority camera; Performing priority analysis and corresponding sorting on the first camera, the second camera and the priority camera to obtain a priority camera list; Filter the first two cameras from the priority camera list as applicable cameras; Sending a video pull request to the applicable camera to obtain real-time road monitoring video; The captured road surveillance video is displayed in real time on the target surveillance screen, and stored and backed up as needed; According to the estimated position of the target vehicle, query the camera database within the set range to obtain the second camera database; Mark all target cameras in the second camera database as pre-selected cameras; According to the first position of the pre-selected camera, all pre-selected cameras are sorted from largest to smallest according to the estimated position of the current target vehicle to obtain a pre-selected camera list; Traversing each pre-selected camera in the pre-selected camera list in turn, deleting the pre-selected camera whose estimated position of the target vehicle is not in the corresponding target shooting area from the pre-selected camera list, and obtaining a new pre-selected camera list; Selecting the first three pre-selected cameras from the new pre-selected camera list as the reserved cameras, and pulling the real-time video streams of the reserved cameras in advance; The first camera, the second camera and the first priority camera are analyzed and sorted accordingly to obtain a priority camera list, including: Calculating a corresponding priority coefficient for the priority camera based on the first angle and the distance from the current target vehicle; The calculation formula of the priority coefficient is as follows: ; In the formula, It is expressed as the priority coefficient of the current priority camera; It is represented as the first angle between the target shooting direction angle of the current priority camera and the real-time heading angle of the current target vehicle; It represents the upper limit of the angle threshold range; It represents the lower limit of the angle threshold range; It is expressed as the influence weight of the angle between the shooting angle and the real-time heading angle of the vehicle on the priority of the analysis camera; It is represented by the real-time distance between the current priority camera and the target vehicle; It is expressed as the sum of the real-time distances between all priority cameras and the target vehicle; It is expressed as the weight of the distance between the camera and the target vehicle on the priority of the analysis camera; Sort all priority cameras from large to small according to their priority coefficients to get the first priority column; Sort all first cameras in ascending order of distance from the current target vehicle to obtain a second priority column; All second cameras are sorted from small to large according to the absolute value of the first angle between the corresponding target shooting direction angle and the real-time heading angle of the current target vehicle to obtain a third priority column; The first priority list, the second priority list and the third priority list are sequentially concatenated to obtain a priority camera list.

2. A road monitoring video selection method based on vehicle position and heading according to claim 1, characterized in that: Obtain the basic information of the target camera and process it to obtain the first key data, including: Extracting a position content description from the basic information of the current target camera, and performing an orientation description on the acquired position content description according to a set standard description rule to obtain a first orientation position; Combine the camera number in the basic information with the first position, and name the head pole of the current target camera according to the set head pole naming rule; The first orientation position of the target camera, the camera pole name and basic information are output as the first key data.

3. The method for selecting road monitoring videos based on vehicle position and heading according to claim 1, characterized in that: According to the first key data, the shooting area of ​​the current target camera is analyzed to obtain the target shooting area, including: Using a preset measurement tool, measure the relative position of the head pole of the current target camera and the preset reference point to obtain first relative position information; Inputting the first relative position information and the first azimuth position of the target camera into a pre-established angle analysis model to obtain a first possible shooting direction angle of the current target camera; Using a set angle measuring device to measure the second possible shooting direction angle of the current target camera pointing to the opposite pole; Combining and analyzing the first possible shooting direction angle and the second possible shooting direction angle, to obtain a target shooting direction angle; The target shooting direction angle is combined with the fixed shooting angle range of the current target camera to calculate and obtain the target shooting range; Determine a target shooting area of ​​the current target camera based on the target shooting range; The target shooting direction angle, target shooting range, shooting direction and target shooting area set are used as key shooting data and filled into the corresponding specific position of the camera database to which the current target camera belongs; The first possible shooting direction angle and the second possible shooting direction angle are combined and analyzed to obtain the target shooting direction angle, including: Compare the first possible shooting direction angle with the second possible shooting direction angle to obtain the absolute difference of the direction angles; When the direction angle absolute difference is less than the set angle difference threshold lower limit, calculating the average value of the first possible shooting direction angle and the second possible shooting direction angle, and outputting it as the target shooting direction angle; When the direction angle absolute difference falls within the set angle difference threshold range, performing a collective adjustment calculation on the first possible shooting direction angle and the second possible shooting direction angle to obtain an adjusted direction angle, and outputting the adjusted direction angle as a target shooting direction angle; The calculation formula of the adjusted direction angle is as follows: ; In the formula, It is expressed as the adjusted direction angle; It is expressed as the first possible shooting direction angle; It is expressed as the second possible shooting direction angle; It is expressed as the measurement error rate of the set angle measurement device; It is expressed as the model test failure rate of the angle analysis model; It is expressed as natural logarithm; e is expressed as a constant with a value of 2.7; When the absolute difference in the direction angle is greater than the set angle difference threshold upper limit, the first possible shooting direction angle and the second possible shooting direction angle are recalculated and combined for analysis until the target shooting direction angle is obtained.

4. The method for selecting road monitoring videos based on vehicle position and heading according to claim 1, characterized in that: Detect the operating status of the target vehicle in real time, and predict the estimated position of the target vehicle at a future time based on the acquired real-time operating status data, including: Use preset sensors to detect the operating status of the target vehicle in real time and obtain real-time operating status data; After pre-processing the operating status data, the operating status data is converted into a unified data format to obtain first status data; Extracting condition data related to the predicted vehicle position from the first condition data and outputting the data as real-time associated data; The real-time associated data is input into a pre-established position prediction model to predict the vehicle position of the current target vehicle at a future time to obtain an estimated position.

5. A road monitoring video selection method based on vehicle position and heading according to claim 4, characterized in that: The real-time associated data refers to the real-time position, real-time speed and real-time heading angle of the target vehicle.

Images