Intersection pedestrian traffic demand detection method based on coordinate positioning and video feedback

Abstract

The intersection pedestrian traffic demand detection method based on coordinate positioning and video feedback comprises the following steps: a matrix position detection area is arranged on each of four corners of a road intersection, and a video feature acquisition component is arranged on each matrix position detection area; a corresponding detection time interval is set based on the sensor coordinate parameters of the initial contact of pedestrians in the matrix position detection area, and a traffic direction determination matrix is formed when the pedestrian passes through the matrix position detection area; the video feature acquisition component synchronously collects the action features and direction features of the pedestrian, iterated functions are used to calculate the elements of the traffic direction determination matrix, and the regression parameters of the traffic direction determination matrix are calculated; the traffic direction and traffic demand of the pedestrian are transmitted to a controller, the lighting period of the green signal of the corresponding direction, i.e., the pedestrian traffic period, is set in combination with the specific number of pedestrians.

Description

TECHNICAL FIELD

[0001] The application relates to a method for detecting intersection pedestrian traffic demand based on coordinate positioning and video feedback. BACKGROUND

[0002] Traffic signal lights are signal lights for guiding traffic operation, which are composed of red lights, green lights and yellow lights; the red light indicates prohibition of traffic, the green light indicates permission of traffic, and the yellow light indicates warning; generally, the traffic signal lights are arranged at road intersections, and all the signal lights are sequentially lighted or extinguished according to a predetermined time node and time interval and a predetermined rule, so as to ensure safe and on-demand traffic of vehicles and pedestrians.

[0003] However, for some special roads, such as internal roads of scenic spots and parks, the control of traffic signal lights is mostly adjusted by control buttons beside the roads; when pedestrians have road communication demand, the control button is pressed, the red signal light on the road is lighted after a delay of about 10 seconds, the timing starts, the vehicles are prohibited from passing through the road, and the pedestrians start to pass through the intersection; after the timing ends, the green signal light is lighted, and the vehicles continue to pass through the road.

[0004] Although the above-mentioned pedestrian intersection passing method can meet the pedestrian traffic demand to some extent, the passing time of the whole process is usually fixedly set; when the pedestrian flow is large, all the pedestrians may not pass through the intersection in one passing period, the safety factor and the passing efficiency cannot be guaranteed, especially for some complex road intersections, the passing direction and the passing demand of the pedestrians cannot be accurately recognized, and most of the pedestrians do not have the consciousness or habit of pressing the button, or the pedestrians press the button in the case of no passing demand; therefore, the normal passing demand of the pedestrians cannot be met, the normal passing order of the vehicles is affected, the traffic jam phenomenon occurs at the road intersection, and even traffic accidents may occur. SUMMARY

[0005] The application provides a method for detecting intersection pedestrian traffic demand based on coordinate positioning and video feedback, which is reasonable in design, based on data interaction of multiple functional modules, adopts a combination of coordinate positioning detection and video feature collection, and sets up matrix type position detection areas on four corners of the road intersection; through signal feedback change and coordinate data change detection of the sensor, and in cooperation with video feature collection, the passing direction and the passing demand of the pedestrians can be accurately determined, the light-off time of the green signal light and the red signal light can be flexibly set according to the specific number of the pedestrians, that is, the passing period of the pedestrians is automatically set, so that all the pedestrians can safely pass through the road intersection in one passing period; and the automatic control mode is adopted to replace the button pressing operation mode of the pedestrians, the operation steps of the pedestrians are simplified, and operation errors are avoided, so that the safety of the road intersection passing is improved, the probability of traffic jam and traffic accidents is reduced, and the problems in the prior art are solved.

[0006] The technical scheme adopted by the present application to solve the above technical problems is:

[0007] The intersection pedestrian traffic demand detection method based on coordinate positioning and video feedback comprises the following steps:

[0008] S1, a matrix position detection area is arranged on each of the four corners of the road intersection, and the matrix position detection area is used to determine the traffic demand and direction of the pedestrians according to the change trend of the sensor coordinate parameters and the change trend of the sensor signals in the detection area; a video feature acquisition component is arranged on each matrix position detection area, and the video feature acquisition component is used to acquire the video features of the pedestrians in the matrix position detection area, so as to accurately identify the number of pedestrians and assist in identifying the traffic direction and demand of the pedestrians;

[0009] S2, taking the initial contact sensor coordinate parameters of the pedestrians in the matrix position detection area as the reference, a corresponding detection time interval is set, and the change trend of the sensor signal parameters and the change trend of the sensor coordinate parameters are detected when the pedestrians pass through the matrix position detection area to form a traffic direction determination matrix combined with the detection time interval;

[0010] S3, the video feature acquisition component synchronously acquires the motion features and orientation features of the pedestrians, iterated functions are used to calculate the elements of the traffic direction determination matrix, the regression parameters of the traffic direction determination matrix are calculated, and then the traffic direction, traffic demand and specific number of pedestrians are accurately determined;

[0011] S4, the traffic direction and demand of the pedestrians are transmitted to the controller, combined with the specific number of pedestrians, the lighting period of the green signal in the corresponding direction, that is, the pedestrian traffic period, is set, and the traffic period is adjusted and corrected through the detection data of the sensor during the pedestrian traffic process, so that all pedestrians can safely pass through the road intersection, and traffic jams and traffic accidents are avoided.

[0012] A matrix position detection area is arranged on each of the four corners of the road intersection, and a video feature acquisition component is arranged on each matrix position detection area, which comprises the following steps:

[0013] S1.1, the specific size of the road intersection is measured, and the setting area of the matrix position detection area is determined according to the specific size of the road intersection;

[0014] S1.2, according to the setting area of the matrix position detection area and the number of sensors, the sensors are uniformly arranged in the matrix position detection area; the sensors are pressure sensors, which are arranged at equal intervals in the form of a square matrix;

[0015] S1.3, establishing a communication electrical connection between the pressure sensor and the video feature acquisition component by using a communication line and a flip-flop, so that the pressure sensor and the video feature acquisition component are synchronized to perform detection and collection actions.

[0016] The video feature acquisition component includes a high-definition camera and a network transmitter. The high-definition camera is used to collect pedestrian passing video features in the matrix position detection area, and the pedestrian passing video features include pedestrian passing action features and pedestrian orientation features, and the passing direction and the number of pedestrians are identified. The network transmitter is used to transmit the collected pedestrian passing video features and sensor signals to the controller.

[0017] The sensor coordinate parameters of the initial contact of the pedestrian in the matrix position detection area are taken as the reference to set the corresponding detection time interval. The sensor signal parameter change trend and the sensor coordinate parameter change trend are detected when the pedestrian passes through the matrix position detection area, and the detection time interval is combined to form a passing direction determination matrix, including the following steps:

[0018] S2.1, n detection time nodes are equally spaced in the detection time interval, denoted as t1, t2, t3, …, tn. n ;

[0019] S2.2, combined with n detection time nodes, n times of sensor coordinate parameter and sensor signal parameter collection are performed. The sensor coordinate parameters contacted by the pedestrian when passing are set as (x n , y n ); when the pedestrian contacts the sensor, the sensor signal parameter is recorded as 1, and when the pedestrian does not contact the sensor, the sensor signal parameter is recorded as 0;

[0020] S2.3, the pedestrian enters the matrix position detection area. The sensor coordinate parameters and the sensor signals are detected and collected at each detection time node based on the initial contact of the pedestrian in the matrix position detection area. The detection element a n =(t n , x n , y n , 1) is recorded;

[0021] S2.4, set the check element b n =(t n+1 , x n , y n , 0) to determine that the pedestrian has left the last detection coordinate. The sensor signal parameter change trend and the sensor coordinate parameter change trend when the pedestrian passes through the matrix position detection area are obtained. The determination element and the check element are combined to form a passing direction determination matrix

[0022] S=(a n, b n )

[0023] Wherein, n is the collection times of sensor coordinate parameters and sensor signal parameters.

[0024] The number of the determination elements is one more than that of the check elements, and under the action of the check elements, the repeated data of the pedestrians can be avoided, the composition of the passing direction determination matrix is simplified, and the accuracy of the passing direction determination matrix calculation is improved.

[0025] The video feature acquisition assembly synchronously acquires the action feature and the orientation feature of the pedestrians, calculates the elements of the passing direction determination matrix by using an iterative function, calculates the regression parameters of the passing direction determination matrix, and then accurately determines the passing direction, the passing demand and the specific number of the pedestrians.

[0026] S3.1, the video feature acquisition assembly determines the passing demand and the approximate passing direction of the pedestrians according to the action feature of the pedestrians, and determines the number of the pedestrians according to the orientation feature of the pedestrians;

[0027] S3.2, the determination element a n =(t n , x n , y n , 1) and the check element b n =(t n+1 , x n , y n , 0) are marked in the matrix position detection area, the change trend of the two elements is analyzed, the passing route of the pedestrians is obtained, and the passing direction and the passing demand of the pedestrians are obtained;

[0028] S3.3, an iterative function is set to calculate the passing direction determination matrix and verify the passing direction and the passing demand of the pedestrians; the iterative function is:

[0029]

[0030] Wherein, m is the iteration times, F0 is the iteration trigger factor, S0 is the iteration parameter standard value, that is, the standard passing direction determination matrix, a n is the determination element, and B is the iteration coefficient.

[0031] S3.4, the specific value of the iteration coefficient is set according to the actual application scene, the iteration is performed 4 times, the regression parameters of the passing direction determination matrix are calculated, and the accuracy of the passing direction determination matrix is verified; when the regression parameters of the passing direction determination matrix are in the interval (0.8, 1.2), it is determined that the passing direction determination matrix meets the use requirement, and the passing route of the pedestrians marked in the matrix position detection area represents the passing direction and the passing demand of the pedestrians.

[0032] The application adopts the above method, determines the passing demand and direction of the pedestrian through the matrix position detection area according to the sensor coordinate parameter variation trend and the sensor signal variation trend in the detection area; collects the passing video features of the pedestrian in the matrix position detection area through the video feature collection component, so as to accurately identify the number of pedestrians and assist in identifying the passing direction and demand of the pedestrian; forms the passing direction judgment matrix by detecting the sensor signal parameter variation trend and the sensor coordinate parameter variation trend in combination with the detection time interval when the pedestrian passes through the matrix position detection area; calculates the passing direction judgment matrix through the iterative function, verifies the passing direction and demand of the pedestrian, and ensures that the obtained passing route of the pedestrian meets the actual use demand, has the advantages of high precision, high efficiency, safety and practicality. BRIEF DESCRIPTION OF DRAWINGS

[0033] Fig. 1 It is a flowchart of the application.

[0034] Fig. 2 It is an application schematic diagram of the application.

[0035] Fig. 3 It is a structural schematic diagram of the matrix position detection area of the application. DETAILED DESCRIPTION

[0036] In order to clearly illustrate the technical features of the present application, the present application will be described in detail below with specific embodiments and in combination with the accompanying drawings.

[0037] As shown in the Figs. 1-3 intersection pedestrian passing demand detection method based on coordinate positioning and video feedback, the detection method comprises the following steps:

[0038] S1, a matrix position detection area is arranged on each of the four corners of the road intersection, and the matrix position detection area is used to determine the passing demand and direction of the pedestrian according to the sensor coordinate parameter variation trend and the sensor signal variation trend in the detection area; a video feature collection component is arranged on each of the matrix position detection areas, and the video feature collection component is used to collect the passing video features of the pedestrian in the matrix position detection area, so as to accurately identify the number of pedestrians and assist in identifying the passing direction and demand of the pedestrian;

[0039] S2, taking the initial contact sensor coordinate parameter of the pedestrian in the matrix position detection area as a reference, a corresponding detection time interval is set, and the sensor signal parameter variation trend and the sensor coordinate parameter variation trend are detected in combination with the detection time interval when the pedestrian passes through the matrix position detection area to form a passing direction judgment matrix;

[0040] S3, the video feature acquisition component synchronously acquires the motion feature and the direction feature of the pedestrian, calculates the elements of the passing direction determination matrix by using an iterative function, calculates the regression parameters of the passing direction determination matrix, and then accurately determines the passing direction, the passing demand and the specific number of the pedestrian;

[0041] S4, the passing direction and the passing demand of the pedestrian are transmitted to the controller, the specific number of the pedestrian is combined, the lighting period of the green signal in the corresponding direction, i.e., the passing period of the pedestrian, is set, and the passing period is adjusted and corrected via the detection data of the sensor during the passing process of the pedestrian, so that all pedestrians can safely pass through the road intersection, and traffic jams and traffic accidents are avoided.

[0042] A matrix position detection area is arranged at each of the four corners of the road intersection, and a video feature acquisition component is arranged on each matrix position detection area, including the following steps:

[0043] S1.1, the specific size of the road intersection is measured, and the setting area of the matrix position detection area is determined according to the specific size of the road intersection;

[0044] S1.2, the sensors are uniformly arranged in the matrix position detection area according to the setting area of the matrix position detection area and the number of the sensors; the sensors are pressure sensors and are arranged at equal intervals in the form of a square matrix;

[0045] S1.3, a communication line and a flip-flop are used to establish a communication electrical connection between the pressure sensor and the video feature acquisition component, so that the pressure sensor and the video feature acquisition component synchronously detect and collect the action.

[0046] The video feature acquisition component includes a high-definition camera and a network transmitter, the high-definition camera is used to collect the pedestrian passing video features in the matrix position detection area, the pedestrian passing video features include the motion action features of the pedestrian and the direction features of the pedestrian, and the passing direction and the number of the pedestrian are identified; the network transmitter is used to transmit the collected pedestrian passing video features and the sensor signal to the controller.

[0047] The sensor coordinate parameters of the pedestrian initially contacting in the matrix position detection area are taken as the reference to set the corresponding detection time interval, the sensor signal parameter change trend and the sensor coordinate parameter change trend are detected when the pedestrian passes through the matrix position detection area, and the detection time interval is combined to form a passing direction determination matrix, including the following steps:

[0048] S2.1, n detection time nodes are equally spaced in the detection time interval, denoted as t1, t2, t3, …, t n ;

[0049] S2.2, n times of sensor coordinate parameter and sensor signal parameter collection are conducted in combination with n detection time nodes, the sensor coordinate parameter contacted by the pedestrian when passing is set as (x n , y n ); when the pedestrian contacts the sensor, the sensor signal parameter is recorded as 1, and when the pedestrian does not contact the sensor, the sensor signal parameter is recorded as 0;

[0050] S2.3, the pedestrian enters the matrix position detection area, and the sensor coordinate parameter initially contacted by the pedestrian in the matrix position detection area is taken as the reference to conduct detection and collection of the sensor coordinate parameter and the sensor signal at each detection time node, recorded as a n =(t n , x n , y n , 1);

[0051] S2.4, a check element b n =(t n+1 , x n , y n , 0) is set to determine that the pedestrian has left the last detection coordinate, the trend of the sensor signal parameter and the trend of the sensor coordinate parameter when the pedestrian passes through the matrix position detection area are obtained, the determination element and the check element are combined to form a passing direction determination matrix

[0052] S=(a n , b n )

[0053] wherein n is the number of times of collection of the sensor coordinate parameter and the sensor signal parameter.

[0054] The number of determination elements is one more than that of check elements, and under the action of the check elements, the repeated data of the pedestrian can be avoided, the formation of the passing direction determination matrix is simplified, and the accuracy of the calculation of the passing direction determination matrix is improved.

[0055] The video feature collection component synchronously collects the motion feature and the orientation feature of the pedestrian, calculates the elements of the passing direction determination matrix by using an iterative function, calculates the regression parameters of the passing direction determination matrix, and then accurately determines the passing direction, the passing demand and the specific number of pedestrians of the pedestrian, including the following steps:

[0056] S3.1, the video feature collection component determines the passing demand and the approximate passing direction of the pedestrian according to the motion feature of the pedestrian, and determines the number of pedestrians of the pedestrian according to the orientation feature of the pedestrian;

[0057] S3.2, the determination element a n =(t n , xn , y n , 1) and the check element b n = (t n+1 , x n , y n , 0) are marked, the trend of the change of the two elements is analyzed, and the pedestrian passing route is obtained, that is, the passing direction and the passing demand of the pedestrian are obtained;

[0058] S3.3, setting an iterative function to calculate the passing direction determination matrix, and verifying the passing direction and the passing demand of the pedestrian; the iterative function is:

[0059]

[0060] wherein m is the iteration number, F0 is an iteration trigger factor, S0 is an iteration parameter standard value, that is, a standard passing direction determination matrix, a n is a determination element, and B is an iteration coefficient;

[0061] S3.4, setting the specific value of the iteration coefficient according to the actual application scene, iterating 4 times, calculating the regression parameter of the passing direction determination matrix, and verifying the accuracy of the passing direction determination matrix; when the regression parameter of the passing direction determination matrix is in the interval (0.8, 1.2), it is determined that the passing direction determination matrix meets the use requirement, and the pedestrian passing route marked in the matrix position detection area represents the passing direction and the passing demand of the pedestrian.

[0062] The working principle of the intersection pedestrian passing demand detection method based on coordinate positioning and video feedback in the embodiment of the application is: based on the data interaction of multiple functional modules, a matrix position detection area is arranged on each of the four corners of the road intersection in a combined manner of coordinate positioning detection and video feature acquisition, through the signal feedback change and coordinate data change detection of the sensor, and in cooperation with the video feature acquisition, the passing direction and the passing demand of the pedestrian can be accurately determined, and the on and off time of the green and red signal lights can be flexibly set according to the specific number of pedestrians, that is, the passing cycle of the pedestrian is automatically set, so that all pedestrians can safely pass through the road intersection in one passing cycle, and the automatic control mode is used to replace the pedestrian button operation mode, thereby simplifying the operation steps of the pedestrian, avoiding operation errors, improving the safety of the road intersection, and reducing the probability of traffic jams and traffic accidents.

[0063] Since the passing time of the internal road of the scenic area or the park is fixed, one passing cycle may not be able to pass all pedestrians when the passenger flow is large, and the passing cycle is too long when the passenger flow is small, which will affect the normal driving of the car on the road, therefore, the detection method of the application can be well applied to the application scene of the internal road of the scenic area or the park.

[0064] In the overall scheme, the detection method comprises the following steps: a matrix position detection area is arranged on each of the four corners of the road intersection, and the matrix position detection area is used to determine the passing demand and direction of the pedestrian according to the change trend of the sensor coordinate parameter and the change trend of the sensor signal in the detection area; a video feature acquisition component is arranged on each matrix position detection area, and the video feature acquisition component is used to acquire the passing video features of the pedestrian in the matrix position detection area, so as to accurately identify the number of pedestrians and assist in identifying the passing direction and passing demand of the pedestrian; the initial contact sensor coordinate parameter of the pedestrian in the matrix position detection area is taken as a reference to set a corresponding detection time interval, and the change trend of the sensor signal parameter and the change trend of the sensor coordinate parameter are detected when the pedestrian passes through the matrix position detection area to form a passing direction determination matrix in combination with the detection time interval; the video feature acquisition component synchronously acquires the action features and direction features of the pedestrian, iterated functions are used to calculate the elements of the passing direction determination matrix, the regression parameters of the passing direction determination matrix are calculated, and then the passing direction, passing demand and specific number of pedestrians are accurately determined; the passing direction and passing demand of the pedestrian are transmitted to the controller, the specific number of pedestrians is combined to set the lighting period of the green signal in the corresponding direction, that is, the passing period of the pedestrian, and the passing period is adjusted and corrected through the detection data of the sensor during the passing process of the pedestrian, so that all pedestrians can safely pass through the road intersection, and traffic jams and traffic accidents are avoided.

[0065] The core innovation of the present application is to combine coordinate positioning detection and video feature acquisition to obtain the change of the sensor coordinate parameter and the signal parameter, combine the time parameter and the data model, and obtain the accurate passing direction and passing demand of the pedestrian through numerical verification, so that the pedestrian can safely and efficiently pass through the road according to the demand.

[0066] Specifically, the matrix position detection area is arranged on each of the four corners of the road intersection, and the video feature acquisition component arranged on each matrix position detection area comprises the following steps: the specific size of the road intersection is measured, and the setting area of the matrix position detection area is determined according to the specific size of the road intersection; the sensors are uniformly arranged in the matrix position detection area according to the setting area of the matrix position detection area and the number of sensors; the sensor is a pressure sensor, and is arranged at equal intervals in the form of a square matrix; a communication line and a flip-flop are used to establish a communication electrical connection between the pressure sensor and the video feature acquisition component, so that the pressure sensor and the video feature acquisition component synchronously perform detection and acquisition actions.

[0067] The matrix position detection area and the video feature acquisition component are configured, the electrical communication is established, and the overall detection method is automatically controlled to provide an accurate passing route for the pedestrian.

[0068] The video feature acquisition component includes a high-definition camera and a network transmitter. The high-definition camera is used to acquire pedestrian traffic video features within a matrix-style position detection area. The pedestrian traffic video features include pedestrian movement features and pedestrian orientation features, identifying the pedestrian's direction of travel and the number of pedestrians. The network transmitter is used to transmit the acquired pedestrian traffic video features and sensor signals to the controller. Under the action of the established network, the video features and sensor signals are interactively judged in real time to help pedestrians safely cross the road.

[0069] The construction of the mathematical model in this application is also an innovation. Based on the sensor coordinate parameters at the initial contact point of the pedestrian within the matrix-type position detection area, a corresponding detection time interval is set. When the pedestrian passes through the matrix-type position detection area, the changing trends of the sensor signal parameters and the sensor coordinate parameters are combined with the detection time interval to construct a direction-of-travel determination matrix. This includes the following steps: n detection time nodes are set at equal intervals within the detection time interval, denoted as t1, t2, t3, ..., t... n Combined with n detection time points, the sensor coordinate parameters and sensor signal parameters are collected n times. The sensor coordinate parameters touched by the pedestrian during passage are set as (x... n y n When a pedestrian touches the sensor, the sensor signal parameter is recorded as 1; when the pedestrian does not touch the sensor, the sensor signal parameter is recorded as 0. When a pedestrian enters the matrix-type position detection area, the sensor coordinate parameters and sensor signals are detected and collected at each detection time node, based on the initial sensor coordinate parameters of the pedestrian within the matrix-type position detection area, and recorded as the judgment element a. n =(t n x n y n , 1); Set the verification element b n =(t n+1 x n y n The algorithm determines that the pedestrian has left the previous detection coordinate. It obtains the trends in sensor signal parameters and sensor coordinate parameters as the pedestrian passes through the matrix-based position detection area. The determination and verification elements are then combined to form the passage direction determination matrix.

[0070] S=(a n b n )

[0071] Where n represents the number of times the sensor coordinate parameters and sensor signal parameters are collected.

[0072] The sensor signals and sensor coordinates are fused via calculation and construction of the passing direction determination matrix to provide data support for determination of the passing route and setting of the operation time of the traffic light, so that the pedestrians can safely pass through the intersection without affecting the driving of the vehicles and avoiding traffic jam and traffic accidents during passing.

[0073] Preferably, the video feature acquisition component synchronously acquires the motion feature and the orientation feature of the pedestrian, calculates the elements of the passing direction determination matrix by using an iterative function, calculates the regression parameters of the passing direction determination matrix, and further accurately determines the passing direction, the passing demand and the specific number of the pedestrians. n n n n n n+1 n n The change trends of the two elements are analyzed to obtain the moving track in the waiting area, and thus the passing direction and the passing demand of the pedestrian are obtained.An iterative function is set to calculate the passing direction determination matrix and verify the passing direction and the passing demand of the pedestrian.

[0074]

[0075] wherein m is the iteration number, F0 is the iteration trigger factor, S0 is the iteration parameter standard value, i.e., the standard passing direction determination matrix, a n is the determination element, and B is the iteration coefficient.According to the actual application scenario, the specific value of the iteration coefficient is set, the regression parameters of the passing direction determination matrix are calculated by iterating 4 times, and the accuracy of the passing direction determination matrix is verified.When the regression parameters of the passing direction determination matrix are in the interval (0.8, 1.2), the passing direction and the passing demand of the pedestrian can be determined.

[0076] It is particularly pointed out that the detection method of the present application can be applied to most road types such as crossroads, T-shaped intersections or irregular roads, and only the corresponding parameters need to be adjusted.

[0077] ​​​​In summary, the intersection pedestrian traffic demand detection method based on coordinate positioning and video feedback in the embodiment of the application is based on data interaction of multiple functional modules, adopts a combination of coordinate positioning detection and video feature collection, sets a matrix type position detection area on each of the four corners of the road intersection, detects signal feedback changes and coordinate data changes, cooperates with video feature collection, can accurately determine the traffic direction and traffic demand of pedestrians, and flexibly sets the on-off time of green and red signal lights according to the specific number of pedestrians, that is, automatically sets the traffic cycle of pedestrians, so as to ensure that all pedestrians can safely pass through the road intersection in one traffic cycle, and uses an automatic control mode to replace the pedestrian button operation mode, simplifies the pedestrian operation steps, avoids operation errors, thereby improving the safety of road intersection traffic, and reducing the probability of traffic jams and traffic accidents.

[0078] The above detailed description cannot be regarded as a limitation on the protection scope of the application, and any alternative improvement or change made to the embodiments of the application by those skilled in the art falls within the protection scope of the application.

[0079] The details not described in the application are well-known to those skilled in the art.

Claims

1. A method for detecting pedestrian crossing demand at intersections based on coordinate positioning and video feedback, characterized in that, The detection method includes the following steps: S1, a matrix-type position detection area is set at each of the four corners of the road intersection. The matrix-type position detection area is used to determine the pedestrian's passage needs and direction of travel based on the changing trends of sensor coordinate parameters and sensor signals within the detection area. Each matrix-type position detection area is equipped with a video feature acquisition component, which is used to collect pedestrian passage video features within the matrix-type position detection area to accurately identify the number of pedestrians and assist in identifying the pedestrian's direction of travel and passage needs. S2, based on the sensor coordinate parameters of the pedestrian's initial contact within the matrix position detection area, a corresponding detection time interval is set. When the pedestrian passes through the matrix position detection area, the changing trend of the sensor signal parameters and the changing trend of the sensor coordinate parameters are combined with the detection time interval to form a passage direction determination matrix. S3, the video feature acquisition component synchronously acquires the action features and orientation features of pedestrians, and uses an iterative function to calculate the elements of the passage direction determination matrix to calculate the regression parameters of the passage direction determination matrix, thereby accurately determining the passage direction, passage demand and specific number of pedestrians; S4 transmits the pedestrian's direction of travel and travel needs to the controller. Based on the specific number of pedestrians, it sets the lighting cycle of the green signal in the corresponding direction, i.e. the pedestrian travel cycle. During the pedestrian's travel, the controller adjusts and corrects the travel cycle based on the detection data from the sensors to ensure that all pedestrians can pass safely at the intersection and avoid traffic jams and accidents.

2. The method for detecting pedestrian crossing demand at intersections based on coordinate positioning and video feedback according to claim 1, characterized in that, A matrix-style position detection area is set up at each of the four corners of the road intersection. Each matrix-style position detection area is equipped with a video feature acquisition component, including the following steps: S1.1 Measure the specific dimensions of the road intersection and determine the area to be set in the matrix position detection zone based on the specific dimensions of the road intersection; S1.2, Based on the area of ​​the matrix position detection region and the number of sensors, the sensors are evenly arranged in the matrix position detection region; the sensors are pressure sensors, arranged at equal intervals in the form of a square matrix. S1.3, a communication electrical connection is established between the pressure sensor and the video feature acquisition component using a communication line and a trigger, so that the pressure sensor and the video feature acquisition component can perform detection and acquisition actions synchronously.

3. The method for detecting pedestrian crossing demand at intersections based on coordinate positioning and video feedback according to claim 2, characterized in that: The video feature acquisition component includes a high-definition camera and a network transmitter. The high-definition camera is used to acquire pedestrian traffic video features within the matrix-style position detection area. The pedestrian traffic video features include pedestrian movement features and pedestrian orientation features, and identify the pedestrian's traffic direction and number of pedestrians. The network transmitter is used to transmit the collected pedestrian traffic video features and sensor signals to the controller.

4. The method for detecting pedestrian crossing demand at intersections based on coordinate positioning and video feedback according to claim 1, characterized in that, Using the sensor coordinate parameters at the initial contact point of a pedestrian within the matrix-type position detection area as a reference, a corresponding detection time interval is set. The process involves combining the changing trends of sensor signal parameters and sensor coordinate parameters with the detection time interval to construct a direction-of-travel determination matrix. This includes the following steps: S2.1, n detection time nodes are set at equal intervals within the detection time interval, denoted as t1, t2, t3, ..., t n ; S2.2, combining n detection time points, collect sensor coordinate parameters and sensor signal parameters n times, and set the sensor coordinate parameters touched by the pedestrian during passage as (x n y n When a pedestrian touches the sensor, the sensor signal parameter is recorded as 1; when the pedestrian does not touch the sensor, the sensor signal parameter is recorded as 0. S2.3, when a pedestrian enters the matrix-type position detection area, the sensor coordinate parameters initially contacted by the pedestrian within the matrix-type position detection area are used as a reference. Sensor coordinate parameters and sensor signals are detected and collected at each detection time point, denoted as decision element a. n =(t n x n y n ,1); S2.4, Set the verification element b n =(t n+1 x n y n The algorithm determines that the pedestrian has left the previous detection coordinate. It obtains the trends in sensor signal parameters and sensor coordinate parameters as the pedestrian passes through the matrix-based position detection area. The determination and verification elements are then combined to form the passage direction determination matrix. S=(a n ,b n ) Where n represents the number of times the sensor coordinate parameters and sensor signal parameters are collected.

5. The method for detecting pedestrian crossing demand at intersections based on coordinate positioning and video feedback according to claim 4, characterized in that: The number of determination elements is one more than that of verification elements. With the help of verification elements, duplicate pedestrian data can be avoided, the composition of the traffic direction determination matrix can be simplified, and the accuracy of the traffic direction determination matrix calculation can be improved.

6. The method for detecting pedestrian crossing demand at intersections based on coordinate positioning and video feedback according to claim 4, characterized in that, The video feature acquisition component synchronously collects pedestrian action features and orientation features. An iterative function is used to calculate the elements of the travel direction determination matrix, determining the regression parameters of the matrix. This allows for accurate determination of the pedestrian's travel direction, travel needs, and the specific number of pedestrians. The process includes the following steps: S3.1, the video feature acquisition component determines the pedestrian's travel needs and approximate travel direction based on the pedestrian's action characteristics, and determines the number of pedestrians based on the pedestrian's orientation characteristics; S3.2, in the matrix position detection area, determine element a n =(t n x n y n 1) and check element b n =(t n+1 x n y n By marking the two elements (0, 0) and analyzing their changing trends, we can obtain the pedestrian route and thus the pedestrian's direction and needs. S3.3, an iterative function is set to calculate the direction determination matrix and verify the pedestrian's direction and travel needs; the iterative function is: Where m is the iteration number, F0 is the iteration trigger factor, S0 is the standard value of the iteration parameter, i.e., the standard travel direction determination matrix, and a n B is the iteration coefficient for determining the element; S3.4, set the specific value of the iteration coefficient according to the actual application scenario, iterate 4 times, calculate the regression parameter of the passage direction determination matrix, and verify the accuracy of the passage direction determination matrix; when the calculated regression parameter of the passage direction determination matrix is ​​within the interval (0.8, 1.2), it can be determined that the passage direction determination matrix meets the usage requirements, and the pedestrian passage route marked in the matrix position detection area represents the pedestrian's passage direction and passage demand.

Images