On-vehicle Sign Retroreflectivity Detection Method and System for Adaptive Slope Roads

Abstract

The present invention discloses a vehicle-mounted sign retroreflective coefficient detection method and system for adaptive ramp roads. When the relative distance between the sign to be detected in the area to be detected and the detection vehicle meets the preset shooting distance, the road inclination angle information at each moment during the driving of the detection vehicle in the area to be detected, which is collected in real time by the inclination angle sensor, is obtained; according to the road inclination angle information at each moment, the road inclination angle in the detection area is solved; the compensation angle of the detection module is calculated according to the road inclination angle, and the electric adjustment structure is controlled according to the compensation angle to adjust the incident angle of the detection module to the target incident angle in real time, so as to obtain the compensated detection module; the compensated detection module is used to take a picture of the traffic sign to be detected, so as to obtain the image of the sign to be recognized; the image of the sign to be recognized is recognized and processed to obtain the retroreflective brightness coefficient corresponding to the traffic sign to be detected, so as to maintain the effectiveness and accuracy of the detection under complex road conditions.

Description

Technical Field

[0001] The present invention relates to the technical field of road detection, and specifically relates to a vehicle-mounted sign retroreflectivity detection method and system for an adaptive slope road. Background Art

[0002] The retroreflectivity of traffic signs is a key evaluation criterion for measuring the safety performance and readability performance of traffic signs. The effective reading distance of road traffic signs is limited. Therefore, the stronger the retroreflective ability of the retroreflective material on the sign surface, the more it can improve the sign brightness, optimize sign recognition, and improve the visual distance.

[0003] In a general vehicle-mounted sign retroreflectivity detection system, when the detection system is in some special road scenarios, due to external physical conditions, the detection effect is poor. For example, in a general vehicle-mounted implementation method, the upper sign detection module installs the light source and camera at a fixed angle. When the detection vehicle is driving on a steep slope with a large slope (including uphill and downhill), the optical axis of the light source cannot form an intersection with the retroreflective body axis of the sign, and may even deviate from the sign, resulting in the light spot area of the light source not being able to effectively irradiate the sign or the camera not being able to capture the effective detection area of the sign, unable to meet the conditions for effectively measuring the sign, and thus may cause incorrect measurement of the sign or missed detection of the sign. Summary of the Invention

[0004] The purpose of the present invention is to provide a vehicle-mounted sign retroreflectivity detection method and system for an adaptive slope road. When the vehicle is driving on a complex road condition with uneven road surface and large road inclination angle and the detection conditions are limited, the road inclination angle is calculated according to the inclination sensor, and the incident angle of the optical axis of the light source of the detection module is adjusted in real time during the driving process of the vehicle, so as not to affect the detection accuracy of the detection module.

[0005] To achieve the above purpose, the present application provides the following solutions:

[0006] In the first aspect, the present application provides a vehicle-mounted sign retroreflectivity detection method for an adaptive slope road, which respectively initializes the horizontal angles between the inclination sensor, the detection module installed on the top of the detection vehicle and the road. When the detection vehicle is driving on the road in the area to be detected, the following steps are included:

[0007] S1. When the relative distance between the sign to be detected in the area to be detected and the detection vehicle meets the preset shooting distance, obtain the road inclination angle information at each moment during the driving process of the detection vehicle in the area to be detected collected by the inclination sensor in real time;

[0008] S2. Solve the road inclination angle in the detection area according to the road inclination angle information at each moment;

[0009] S3. Calculate the compensation angle of the detection module according to the road inclination angle, and control the electric adjustment structure according to the compensation angle to adjust the incident angle of the detection module to the target incident angle in real time, so as to obtain a compensated detection module;

[0010] S4. Use the compensated detection module to take a picture of the traffic sign to be detected, so as to obtain an image of the sign to be recognized;

[0011] S5. Perform recognition processing on the image of the sign to be recognized to obtain the retroreflective brightness coefficient corresponding to the traffic sign to be detected.

[0012] In some specific implementation schemes, the process of initializing the inclination sensor and the horizontal included angle between the detection module and the road is as follows:

[0013] When the detection vehicle is in a stationary state, adjust the installation angle of the inclination sensor horizontally installed on the detection vehicle to be parallel to the road, and at the same time adjust the incident angle to the target incident angle. The incident angle is the included angle between the light axis of the light source of the detection module at the top of the detection vehicle and the retroreflective body axis parallel to the road of the traffic sign to be detected above the road.

[0014] In some specific implementation schemes, the road inclination information includes pitch angle measurement values. The specific process of step S2 is as follows:

[0015] S21. Set the sampling period of the road inclination information, and store the n pitch angle measurement values collected in each sampling period into a collection of acquisition data, denoted as X (X1,..., Xn);

[0016] S22. Perform a first-order filtering process on each pitch angle measurement value in the collection of acquisition data to obtain n preprocessed data;

[0017] S23. Use studentized residuals to screen the n preprocessed data again, remove the outliers with too large deviations, and then calculate the average value of the remaining data to obtain the optimal estimated value of the road inclination angle. Take the optimal estimated value of the road inclination angle as the road inclination angle of the detection vehicle in the current area to be detected.

[0018] In some specific implementation schemes, the specific process of the first-order filtering process in step S22 is as follows:

[0019] Input the pitch angle measurement values in the collection of acquisition data into the first-order filter for preprocessing in sequence according to the acquisition time. The preprocessing process is to weight the pitch angle measurement value at the current acquisition time and the filtered output value filtered by the first-order filter at the previous acquisition time to obtain the filtered output value at the current acquisition time. Take the filtered output values at each acquisition time as the preprocessed data;

[0020] The calculation method of the first-order filter is as follows:

[0021]

[0022] Among them, α represents the filtering coefficient, represents the pitch angle measurement value at the current acquisition moment, represents the filtered output value at the previous acquisition moment, represents the filtered output value at the current acquisition moment.

[0023] In some specific implementation manners, the specific process of step S23 is as follows:

[0024] S231. Calculate the average value of n preprocessed data and the residual corresponding to each preprocessed data;

[0025] S232. Calculate the standard deviation of n preprocessed data according to the residual of each preprocessed data;

[0026] S233. Calculate the studentized residual corresponding to each preprocessed data according to the standard deviation and the residual of each preprocessed data;

[0027] S234. Traverse n preprocessed data, and respectively determine whether the studentized residual corresponding to each preprocessed data exceeds a preset threshold. If so, remove the preprocessed data. If not, save it to the road inclination angle dataset separately;

[0028] S235. Calculate the average value of the data in the road inclination angle dataset to obtain the optimal estimated value of the road inclination angle.

[0029] In some specific implementation manners, the electric adjustment structure is used to drive the light source optical axis to move. The specific process of step S3 is as follows:

[0030] S31. Obtain the actual included angle between the light source optical axis of the detection module and the horizontal plane according to the sum of the compensation angle and the target incident angle;

[0031] S32. Input the actual included angle and the target incident angle into the established mathematical model of the movement relationship between the travel distance of the electric adjustment structure and the road inclination angle, and calculate the movement distance a of the electric adjustment structure;

[0032] S33. Judge the ramp direction in which the detection vehicle travels according to the road inclination angle, and control the electric adjustment structure to adjust the incident angle of the light source optical axis to the target incident angle according to the ramp direction and the movement distance a.

[0033] In some specific implementation manners, the electric adjustment structure includes a stepper motor with a lead screw and a slider that moves along with the lead screw and drives the light source optical axis to move. The mathematical model of the movement relationship in step S32 is:

[0034]

[0035] Wherein, L represents the length of the light source mounting plate with the mounting direction parallel to the optical axis of the light source, ω represents the actual angle, and β represents the target incident angle. represents the minimum acute angle of the slider cross-section.

[0036] In some specific embodiments, in step S33, the process of judging the ramp direction of the detection vehicle according to the road inclination is as follows:

[0037] When the road inclination is positive, the ramp direction is uphill; when the road inclination is negative, the ramp direction is downhill. When the detection vehicle is traveling uphill, the lead screw drives the slider to move forward a distance a in the direction of the detection vehicle's travel; when the detection vehicle is traveling downhill, the lead screw drives the slider to move backward a distance a in the direction opposite to the detection vehicle's travel.

[0038] In a second aspect, the present application provides a vehicle-mounted sign retroreflective coefficient detection system for adapting to sloped roads, including an observation module, a detection module, an inclination sensor, and an electric adjustment structure for adjusting the incident angle of the detection module, which are respectively communicatively connected to a control module. Among them,

[0039] The observation module includes a camera and a radar. The optical axes of the camera and the radar are installed parallel to the length direction of the detection vehicle on the top of the detection vehicle for taking images of the area to be detected;

[0040] The detection module includes a detection camera, a light source for illumination, and a mounting bracket for fixing the detection camera and the light source, and adjusts the angle between the optical axis of the light source and the retroreflector axis of the traffic sign to be detected to the target incident angle;

[0041] The inclination sensor is horizontally installed on the top of the detection vehicle for measuring the road inclination information at any time when the detection vehicle is traveling on the road;

[0042] The electric adjustment structure is installed on the roof rack of the detection vehicle and is movably connected to the mounting bracket, and is used to drive the mounting bracket to move under the control of the control module, so that the incident angle between the optical axis of the light source and the retroreflector axis of the traffic sign to be detected remains unchanged;

[0043] The control module is used to execute the vehicle-mounted sign retroreflective coefficient detection method described in the first aspect.

[0044] In some specific embodiments, the electric adjustment structure includes a stepper motor with a lead screw and an encoder. A lead screw guide rail is fixed on the top of the detection vehicle. A wedge-shaped slider is installed on the lead screw guide rail. The cross-section of the wedge-shaped slider is an isosceles right triangle with an acute angle of σ. The rotation of the lead screw drives the wedge-shaped slider to reciprocate along the lead screw guide rail. The wedge-shaped slider is connected to one end of the mounting bracket through a bearing, thereby raising or lowering the incident angle of the optical axis of the light source.

[0045] The inventive concept of the present application is:

[0046] When the vehicle is driving on complex road conditions with uneven roads and large road inclinations, it is very likely that the optical axis of the light source of the detection module cannot form an intersection with the retroreflector axis of the traffic sign, and may even deviate from the sign, thus failing to meet the effective shooting of the sign and affecting the detection of the retroreflective coefficient of the sign. Therefore, when using on-vehicle retroreflective coefficient detection, it is required that when the detection vehicle is driving on a straight road and reaches the shooting point, the angle between the optical axis of the light source of the detection module above the detection vehicle and the retroreflector axis of the sign must form a specific angle. This angle is called the incident angle β, and the incident angle needs to remain unchanged, with the absolute value of β being 4 degrees. The observation angle α (the angle between the optical axis of the light source and the optical axis of the camera) is determined by the installation angle between the light source and the camera during installation, and the observation angle is 0.2 degrees.

[0047] Therefore, in order to keep the incident angle unchanged and meet the detection requirements of the detection module in this application, when the detection vehicle is driving on the road to be detected, the road inclination of the current driving road of the detection vehicle is calculated by the inclination sensor. If the road inclination exceeds the set threshold, the electric adjustment structure is driven to lift the detection module above to compensate for the corresponding inclination, so that the optical axis of the light source can always have an intersection with the retroreflector axis of the sign, and the absolute value of the incident angle β formed between them always remains 4 degrees.

[0048] The beneficial effects of the present invention compared with the prior art:

[0049] The present invention takes into account that the road slope often changes during the driving process of the detection vehicle, resulting in inconsistent conditions for measuring the retroreflective coefficient of the sign. Based on the measurement of the road inclination by the inclination sensor, it can adjust the angle between the detection module and the retroreflector axis of the sign in real time, ensure the consistency of the shooting conditions, reduce the false detection and missed detection of the sign, and improve the comprehensiveness of the sign detection and the adaptability to different complex working conditions; aiming at the detection vehicle can still maintain the effectiveness and accuracy of detection under complex road conditions with uneven roads and large road inclinations.

[0050] The present invention is suitable for high vehicle speeds and large slopes, has a high measurement frequency, a fast adjustment speed, good real-time performance, and is suitable for continuous slope sections. And it adjusts the incident angle in real time during the detection process, without the need to decelerate or stop, and does not affect the detection. It is proposed to use a first-order low-pass filter and studentized residuals for data cleaning to effectively avoid measurement deviations and jump errors. Description of the Drawings

[0051] Figure 1 It is the structural block diagram of the on-vehicle sign retroreflective coefficient detection system for an adaptive slope road provided by the embodiment of the present invention;

[0052] Figure 2 It is the flow chart of the on-vehicle sign retroreflective coefficient detection method for an adaptive slope road provided by the embodiment of the present invention;

[0053] Figure 3 Side view of a straight road provided by an embodiment of the present invention;

[0054] Figure 4 Side view of a ramp road provided by an embodiment of the present invention;

[0055] Figure 5 Schematic diagram of a mathematical model provided by an embodiment of the present invention;

[0056] Figure 6 Schematic diagram of the mathematical model after the slider slides provided by an embodiment of the present invention. Detailed implementation manners

[0057] Next, the technical solutions in the embodiments of the present invention will be clearly and completely described in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only a part of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is actually only illustrative and in no way limits the present invention and its application or use. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative efforts belong to the scope of protection of the present invention.

[0058] Unless otherwise specifically stated, the relative arrangements of components and steps, numerical expressions and values described in these embodiments do not limit the scope of the present invention.

[0059] At the same time, it should be understood that, for the sake of description, the sizes of the various parts shown in the drawings are not drawn according to the actual proportional relationship.

[0060] In addition, for the sake of clarity and conciseness, the descriptions of well-known structures, functions and configurations may be omitted. Those of ordinary skill in the art will recognize that various changes and modifications can be made to the examples described herein without departing from the spirit and scope of the present disclosure.

[0061] The technologies, methods and devices known to those of ordinary skill in the relevant fields may not be discussed in detail, but where appropriate, the technologies, methods and devices should be regarded as part of the authorization specification.

[0062] In all the examples shown and discussed here, any specific value should be construed as merely exemplary, rather than as a limitation. Therefore, other examples of the exemplary embodiments may have different values.

[0063] Embodiment 1

[0064] As Figure 1As shown in the figure, this embodiment provides a vehicle-mounted reverse reflectance brightness coefficient detection system, which includes a mileage measurement module, an observation module, a detection module, an inclination sensor, and an electric adjustment structure for adjusting the incident angle of the detection module, all of which are communicatively connected to the control module. Among them,

[0065] The observation module includes a camera and a radar. The optical axes of the camera and the radar are installed parallel to the length direction of the detection vehicle on the top of the detection vehicle for taking images of the area to be detected.

[0066] The detection module includes a detection camera, a light source for illumination, and a mounting bracket for fixing the detection camera and the light source, and adjusts the angle between the optical axis of the light source (i.e., the light source optical axis) and the retroreflector axis of the traffic sign to be detected to the target incident angle; the detection module in this embodiment specifically refers to the upper sign detection module for photographing the traffic signs above the road area to be detected; the camera and the light source in the upper sign detection module are fixed on the mounting bracket at a determined interval and angle. In this embodiment, the angle between the optical axis of the camera and the optical axis of the light source is 0.2 degrees, and the mounting bracket is installed on the electric adjustment structure.

[0067] The inclination sensor is horizontally installed on the top of the detection vehicle and is used to measure the road inclination information at any time when the detection vehicle is driving on the road.

[0068] The inclination sensor is horizontally installed on the detection vehicle. The inclination sensor includes a high-precision gyroscope, an accelerometer, and a geomagnetic sensor. The gyroscope collects the angular velocities of the three axes of the vehicle, and the accelerometer collects the accelerations of the three axes of the vehicle. The inclination sensor can directly measure information such as the angular velocities of the three axes, the accelerations of the three axes, the pitch angle, and the heading angle of the detection vehicle at any time. The integration of multiple sensors can greatly overcome the errors and uncertainties caused by a single sensor. The longitudinal detection accuracy of the inclination sensor is 0.001 degrees, that is, the pitch angle detection accuracy in the longitudinal direction of the detection vehicle is 0.001 degrees, and the feedback rate is 200 hHZ, which can adapt to a rapidly changing dynamic environment. The longitudinal inclination detection range is plus or minus 90 degrees.

[0069] The electric adjustment structure is installed on the roof rack of the detection vehicle and is movably connected to the mounting bracket, and is used to drive the mounting bracket to move under the control of the control module, so that the incident angle between the optical axis of the light source and the retroreflector axis of the traffic sign to be detected remains unchanged.

[0070] The electric adjustment structure includes a stepper motor with a lead screw and an encoder. A lead screw guide rail is fixed on the top of the detection vehicle. A wedge-shaped slider is installed on the lead screw guide rail. The cross-section of the wedge-shaped slider is an isosceles right triangle with an acute angle of σ. The rotation of the lead screw drives the wedge-shaped slider to reciprocate along the lead screw guide rail. The wedge-shaped slider is connected to one end of the mounting bracket through a bearing, thereby raising or lowering the incident angle of the light source optical axis.

[0071] The mileage measurement module is installed on the left rear wheel hub of the inspection vehicle to measure the road mileage of the inspection vehicle during driving;

[0072] The control module is used to execute an on-vehicle sign retroreflectivity detection method for an adaptive slope road described in the first aspect.

[0073] The control module executes an on-vehicle sign retroreflectivity detection method for an adaptive slope road according to data such as the image and point cloud data of the area to be detected, the sign to be detected, the road mileage of the inspection vehicle, and the road inclination information received from each module.

[0074] Embodiment 2

[0075] As Figure 2 shown, based on the measurement system carried in Embodiment 1 Figure 1 This embodiment provides an on-vehicle sign retroreflectivity detection method for an adaptive slope road. The measurement method specifically includes the following steps:

[0076] Step 1: Install an inclination sensor on the inspection vehicle and adjust the installation position

[0077] Initialize the horizontal angles between the inclination sensor installed on the top of the inspection vehicle, the detection module and the road respectively. When the inspection vehicle is in a stationary state, adjust the installation angle of the inclination sensor horizontally installed on the inspection vehicle to be parallel to the road, and at the same time adjust the incident angle to the target incident angle. The incident angle is the angle between the light axis of the light source of the detection module on the top of the inspection vehicle and the retroreflector axis parallel to the road of the traffic sign to be detected above the road. Specifically, park the inspection vehicle with the engine off on a horizontal road surface, and install the inclination sensor horizontally on the inspection vehicle. According to the measured installation pitch angle and installation roll angle, adjust the installation angle of the inclination sensor until both the current installation pitch angle and installation roll angle are zero. After the static adjustment is completed, drive the inspection vehicle to travel at a constant speed on a straight and horizontal road, and adjust the installation yaw angle in real time until it is zero.

[0078] The installation pitch angle refers to the angle between the inclination sensor (along the length direction of the vehicle) and the ground (horizontal plane), the installation roll angle refers to the angle between the inclination sensor (along the width direction of the vehicle) and the ground (horizontal plane), and the yaw angle refers to the angle between the front of the inclination sensor and the inspection vehicle (along the length direction of the vehicle).

[0079] Step 2: When the inspection vehicle is driving on the road in the area to be inspected, the inclination sensor is used to collect the road inclination information of the vehicle at different times in real time. When the inspection vehicle is driving on the inspection road, the inclination sensor detects the inclination θ of the current road at a high frequency and continuously sends it to the control module. When the inspection vehicle is moving at a constant speed or stationary on a horizontal road surface, θ is 0 degrees. However, due to unevenness, potholes, etc. on the road and a certain degree of bumps and vibrations during the driving of the vehicle, the output of the inclination sensor will show a certain degree of fluctuation, and the sampled values need to be preprocessed first. The specific steps are as follows:

[0080] S1. When the relative distance between the sign to be inspected in the area to be inspected and the inspection vehicle meets the preset shooting distance, obtain the road inclination information of the inspection vehicle at each moment during the driving process in the area to be inspected collected by the inclination sensor in real time;

[0081] The relative distance between the sign to be inspected and the inspection vehicle is obtained by using the image data and point cloud data of the road area to be inspected captured by the observation module on the top of the inspection vehicle. The image data of the road area to be inspected is input into a pre-trained first deep learning detection model, and combined with the point cloud data, the relative distance between the sign to be inspected in the road area to be inspected and the inspection vehicle is obtained; The first deep learning detection model uses a yolov8-based object detection model, and its lightweight and real-time performance can ensure good detection effects and detection speeds under resource-constrained conditions.

[0082] When the inspection vehicle is driving on the road to be inspected, the radar and camera of the observation module capture the area to be inspected at a fixed frequency. The fixed frequency can be freely set by the control module to trigger by distance or time interval. In this embodiment, fixed time interval triggering is used for acquisition, and the acquisition interval time is 200 ms. When the inspection vehicle is driving on the road to be inspected, the control module triggers the camera and radar in the observation module to capture the two-dimensional image data and point cloud data of the road area to be inspected at a fixed frequency simultaneously.

[0083] When the relative distance between the sign to be inspected in the area to be inspected and the inspection vehicle meets the preset shooting distance, the detection module is triggered to take continuous photos of the sign to be inspected with and without light sources, and two consecutive frame images are obtained for subsequent recognition. Since the photos taken need to satisfy that the optical axis and the inverse body axis have an intersection point, the following steps are performed:

[0084] S2. According to the road inclination information at each moment, solve the road inclination in the inspection area;

[0085] The road inclination information includes pitch angle measurement values. The specific process of step S2 includes:

[0086] S21. Set the sampling period of the road inclination information, and store the n pitch angle measurement values collected in each sampling period into a collection of acquisition data, denoted as X (X1, …, Xn);

[0087] S22. Perform a first-order filtering process on each pitch angle measurement value in the collection of acquisition data to obtain n preprocessed data;

[0088] The specific process of the first-order filtering process in step S22 is as follows:

[0089] Input the pitch angle measurement values in the collection of acquisition data into the first-order filter for preprocessing in sequence according to the acquisition time. The preprocessing process is to weight the pitch angle measurement value at the current acquisition time and the filtered output value filtered by the first-order filter at the previous acquisition time to obtain the filtered output value at the current acquisition time, and use the filtered output values at each acquisition time as the preprocessed data;

[0090] The calculation method of the first-order filter is:

[0091]

[0092] where, represents the filtering coefficient. The smaller the filtering coefficient , the smoother the result, the better the anti-interference performance, and it can effectively remove the abnormal angle values caused by the rapid change of the inclination angle when the detection vehicle bumps on the road and during acceleration and deceleration. represents the pitch angle measurement value at the current acquisition time, represents the filtered output value at the previous acquisition time, represents the filtered output value at the current acquisition time.

[0093] When the detection vehicle is traveling on the detection road, the inclination angle sensor detects the inclination angle θ of the current road at a high frequency and continuously sends it to the control module. When the detection vehicle is moving at a constant speed or stationary on a horizontal road surface, θ is 0 degrees. However, due to the unevenness and potholes of the road and the certain degree of bumps and vibrations during the driving of the vehicle, the output of the inclination angle sensor will show a certain degree of fluctuation, and it is necessary to preprocess the sampled values first. Therefore, in this embodiment, based on the pitch angle measurement values of the inclination angle sensor, the pitch angle data within a certain time period is selected, and its average value is used as the road inclination angle of this section of the test area.

[0094] Set every 100 ms as a sampling period. n pitch angle values can be obtained in one period. First, use a first-order filter to preprocess the data, filter out high-frequency interference, and weight the current sampling value and the previous filtered output value to obtain the current effective filtered value. The sampling time length can be set according to specific needs.

[0095] S23. Use the studentized residuals to screen the n preprocessed data again. After removing the outliers with too large deviations, calculate the average of the remaining data to obtain the optimal estimated value of the road inclination angle, and use the optimal estimated value of the road inclination angle as the road inclination angle of the detection vehicle in the current area to be detected.

[0096] The specific process of step S23 is as follows:

[0097] S231. Calculate the average of the n preprocessed data and the residual corresponding to each preprocessed data;

[0098] S232. Calculate the standard deviation of the n preprocessed data according to the residual of each preprocessed data;

[0099] S233. Calculate the studentized residual corresponding to each preprocessed data according to the standard deviation and the residual of each preprocessed data;

[0100] S234. Traverse the n preprocessed data, and respectively determine whether the studentized residual corresponding to each preprocessed data exceeds a preset threshold. If so, remove the preprocessed data. If not, save it to the road inclination angle dataset separately;

[0101] S235. Calculate the average of the data in the road inclination angle dataset to obtain the optimal estimated value of the road inclination angle.

[0102] Calculate the average of the preprocessed data and the corresponding residuals, and use the studentized residuals to reliably detect outliers to prevent these outliers from having an adverse impact on the final result. The studentized residuals can more sensitively capture outliers and avoid the "dilution" of outliers on the final result. For the n data obtained after preprocessing, calculate the average , and the formula is: , where represents the i-th angle value after preprocessing.

[0103] Calculate the residual , and the formula is .

[0104] Calculate the standard deviation s , and the formula is .

[0105] Calculate the studentized residual , and the formula is .

[0106] If the absolute value of the studentized residual is large, it means that this observation value may be an outlier. Remove the outliers with too large deviations, and then calculate the average of the remaining data. Through the above method, obtain the optimal estimated value of the road inclination angle, and use the optimal estimated value as the inclination angle θ of the road where the detection vehicle is currently located.

[0107] S3. Calculate the compensation angle of the detection module according to the road inclination angle, and control the electric adjustment structure to adjust the incident angle of the detection module to the target incident angle in real time, so as to obtain the compensated detection module.

[0108] According to the calculated road inclination angle θ, the motor is driven in real time to adjust the compensation angle of the upper detection module. The compensation angle of the upper detection module is the road inclination angle θ. The incident angle before angle compensation is β, and after adjustment, the incident angle formed by the light source optical axis and the retroreflector axis of the sign can still be β. The electric adjustment structure is used to drive the movement of the light source optical axis. The specific process of step S3 is as follows:

[0109] S31. Obtain the actual angle between the light source optical axis of the detection module and the horizontal plane according to the sum of the compensation angle and the target incident angle.

[0110] S32. Input the actual angle and the target incident angle into the established mathematical model of the movement relationship between the travel distance of the electric adjustment structure and the road inclination angle, and calculate the moving distance a.

[0111] The electric adjustment structure includes a stepper motor with a lead screw, and a slider that moves along with the lead screw and drives the movement of the light source optical axis. The mathematical model of the movement relationship in step S32 is:

[0112]

[0113] where L represents the length of the light source mounting plate with the installation direction parallel to the light source optical axis, ω represents the actual angle, β represents the target incident angle, represents the minimum acute angle of the slider cross-section.

[0114] S33. Judge the ramp direction of the detection vehicle according to the road inclination angle, and control the lead screw of the stepper motor to drive the slider to move a distance a in the corresponding direction according to the ramp direction and the moving distance a, so that the incident angle of the light source optical axis is adjusted to the target incident angle.

[0115] In step S33, the process of judging the ramp direction of the detection vehicle according to the road inclination angle is as follows:

[0116] When the road inclination angle is positive, the ramp direction is uphill; when the road inclination angle is negative, the ramp direction is downhill. When the detection vehicle is traveling uphill, the lead screw drives the slider to move forward a distance a in the direction of the detection vehicle's travel; when the detection vehicle is traveling downhill, the lead screw drives the slider to move backward a distance a in the direction opposite to the detection vehicle's travel.

[0117] In this embodiment, the stepper motor is a stepper motor with a ball screw and an encoder. The slider is a wedge-shaped slider. The maximum speed of the screw can reach 100 mm / s, and the repeat positioning accuracy is less than 0.01 mm. The encoder is a 12-bit, 2000-line absolute encoder. After obtaining the road inclination angle θ, the operation module calculates the stroke that the screw needs to move corresponding to the light source compensation angle, and drives the motor to complete the compensation of the light source optical axis angle.

[0118] As Figures 5 - 6 shown, a wedge-shaped slider is slidably connected to the screw guide rail. The cross-section of the wedge-shaped slider is an isosceles right triangle with an acute angle of σ. The rotation of the screw drives the wedge-shaped slider to move back and forth along the linear guide rail. The wedge-shaped slider is connected to one end of the light source bracket through a bearing, thereby lifting or lowering the inclination angle of the light source.

[0119] Therefore, it is necessary to solve the relationship between the angle of lifting or lowering of the light source and the stroke of the screw: Simplify the above problem into a mathematical model, and consider the case where the angle β between the light source optical axis and the retroreflector axis of the signboard is 4 degrees at the initial moment. The initial state is as Figure 5 shown. When the road inclination angle is θ, the state after the motor drives the screw / slider to travel a distance of a is as Figure 6 .

[0120] The following relationships exist between the variables:

[0121] .

[0122] The relationship with the travel distance of the screw is

[0123]

[0124]

[0125]

[0126] The travel distance of the screw is obtained a The relationship with the road inclination angle is:

[0127] Among them, represents the height difference generated when the slider travels; represents the height of the intersection point of the light source mounting plate and the wedge-shaped slider from the fixed point of the light source mounting plate at the initial moment; It represents the height of the intersection point between the light source mounting plate and the wedge-shaped slider from the fixed point of the light source mounting plate after the slider travels a distance a; since the light source is fixed on the light source mounting plate, the mounting direction of the light source mounting plate is parallel to the optical axis of the light source. Therefore, the angle between the light source mounting plate and the horizontal plane is the same as the angle between the optical axis of the light source and the horizontal plane. a represents the distance that the slider travels along the guide rail; σ represents the minimum acute angle of the cross-section of the wedge-shaped slider; β represents the initial incident angle; ω represents the actual angle between the light source mounting plate and the horizontal plane after angle compensation; L represents the length of the light source mounting plate. When going uphill, θ is positive, and when going downhill, θ is negative. The sign is used to distinguish between the two states of going uphill and downhill. Without special instructions, the θ substituted into the formula in this article refers to the positive value or the absolute value.

[0128] The distance a that the lead screw needs to travel is calculated from the compensation angle. When the inspection vehicle is traveling uphill, the motor drives the lead screw to move forward a distance a to reduce the inclination of the optical axis of the light source relative to the horizontal plane; as Figure 4 shown, when the inspection vehicle is traveling downhill, the motor drives the lead screw to move backward a distance a to increase the inclination of the optical axis of the light source relative to the horizontal plane; as Figure 3 shown, when the inspection vehicle is traveling on a horizontal road surface, the motor drives the lead screw back to the initial state.

[0129] According to standards such as the "Technical Standards for Highway Engineering", the road inclination angle generally does not exceed 10°.

[0130] The stroke of the lead screw within 100 ms can reach 10 mm. In the present invention, L = 150 mm and σ = 45°. According to the above formula, it can be calculated that the maximum angle range that can be changed within 100 ms can reach approximately 4°. It can dynamically adjust the angle between the optical axis of the light source and the axis of the retroreflector of the sign in real time before the next shooting without affecting the next shooting.

[0131] S4. Use the compensated detection module to take a picture of the traffic sign to be detected to obtain an image of the sign to be recognized;

[0132] S5. Perform recognition processing on the image of the sign to be recognized to obtain the retroreflective brightness coefficient corresponding to the traffic sign to be detected.

[0133] The focus of this application is on how to adaptively adjust the situation where the optical axis of the light source cannot form an intersection with the axis of the retroreflector of the sign during the detection process of the sign retroreflective coefficient, and may even deviate from the sign, unable to meet the conditions for effectively measuring the sign, which may cause incorrect measurement of the sign or missed detection of the sign. The specific recognition method for the retroreflective coefficient is not limited.

[0134] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can take the form of a complete hardware embodiment, a complete software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can take the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk memory, CD-ROM, optical memory, etc.) that contain computer-usable program code.

[0135] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be realized by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0136] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory generate a manufactured article including instruction means that realize the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0137] These computer program instructions can also be loaded onto a computer or other programmable data processing device, and a series of operation steps are executed on the computer or other programmable device to generate a computer-implemented process, so that the instructions executed on the computer or other programmable device provide steps for realizing the functions specified in Figure 1 one flow or multiple flows and / or blocks Figure 1 one block or multiple blocks.

[0138] As mentioned above, it is only the preferred embodiments of the present invention, and there is no any formal limitation to the present invention. According to the technical essence of the present invention, any simple modification, equivalent replacement, and improvement made to the above embodiments within the spirit and principle of the present invention still fall within the protection scope of the technical solution of the present invention.

Claims

1. A method for detecting the retroreflection coefficient of a vehicle-mounted sign on an adaptive slope road, characterized in that: Initialize the tilt sensor installed on the top of the inspection vehicle, the horizontal angle between the inspection module and the road respectively. When the inspection vehicle is driving on the road in the area to be inspected, the following steps are included: S1. When the relative distance between the sign to be detected in the area to be detected and the detection vehicle meets the preset shooting distance, the road inclination information at each moment during the driving process of the detection vehicle in the area to be detected collected in real time by the inclination sensor is obtained; S2. Calculate the road inclination angle in the detection area according to the road inclination angle information at each moment; S3, calculating the compensation angle of the detection module according to the road inclination angle, and controlling the electric adjustment structure according to the compensation angle to adjust the incident angle of the detection module to the target incident angle in real time, so as to obtain a compensated detection module; The electric adjustment structure is used to drive the light source optical axis to move. The electric adjustment structure includes a stepping motor with a lead screw and a slider that moves with the lead screw and drives the light source optical axis to move. The specific process of step S3 is: S31, obtaining the actual angle between the light source axis of the detection module and the horizontal plane according to the sum of the compensation angle and the target incident angle; S32, inputting the actual angle and the target incident angle into the established mathematical model of the motion relationship between the travel distance of the electric adjustment structure and the road inclination angle, and calculating the moving distance a of the electric adjustment structure; The motion relationship mathematical model in step S32 is: ； Wherein, L represents the length of the light source mounting plate whose mounting direction is parallel to the optical axis of the light source, ω represents the actual angle, β represents the target incident angle, and σ represents the minimum acute angle of the slider cross section; S33, judging the ramp direction of the inspection vehicle according to the road inclination angle, and controlling the electric adjustment structure to adjust the incident angle of the light source optical axis to the target incident angle according to the ramp direction and the moving distance a; In step S33, the process of determining the direction of the slope on which the detection vehicle is traveling according to the road inclination angle is as follows: When the road inclination is positive, the ramp direction is uphill, and when the road inclination is negative, the ramp direction is downhill. When the inspection vehicle is traveling uphill, the screw drives the slider to move forward a distance a in the direction of the inspection vehicle. When the inspection vehicle is traveling downhill, the screw drives the slider to move backward a distance a in the direction opposite to the inspection vehicle. S4, using the compensated detection module to take a photo of the traffic sign to be detected to obtain an image of the sign to be identified; S5. Perform image recognition processing on the traffic sign to be detected to obtain the retroreflection brightness coefficient corresponding to the traffic sign to be detected.

2. The method for detecting the retroreflection coefficient of a vehicle-mounted sign on an adaptive slope road according to claim 1, characterized in that: The process of initializing the tilt sensor and detecting the horizontal angle between the module and the road is as follows: When the inspection vehicle is stationary, the installation angle of the tilt sensor installed horizontally on the inspection vehicle is adjusted to be parallel to the road, and the incident angle is adjusted to the target incident angle, which is the angle between the optical axis of the light source of the detection module on the top of the inspection vehicle and the axis of the retroreflector of the traffic sign to be inspected above the road and parallel to the road.

3. The method for detecting the retroreflection coefficient of a vehicle-mounted sign on an adaptive slope road according to claim 2, characterized in that: The road inclination information includes a pitch angle measurement value. The specific process of step S2 includes: S21, set the sampling period of the road inclination information, the n pitch angle measurement values ​​collected in each sampling period are stored in a collection data set, denoted as X (X1, ..., Xn); S22, using a first-order filter to perform a filtering process on each pitch angle measurement value in the collected data set to obtain n pre-processed data; S23. Filter the n preprocessed data again using the studentized residual, remove the outliers with excessive deviations, average the remaining data, and obtain the optimal estimated value of the road inclination. The optimal estimated value of the road inclination is used as the road inclination of the detection vehicle in the current area to be detected.

4. The method for detecting the retroreflection coefficient of a vehicle-mounted sign on an adaptive slope road according to claim 3, characterized in that: The specific process of the primary filtering process in step S22 is as follows: The pitch angle measurement values ​​in the collected data set are sequentially input into the first-order filter for preprocessing according to the collection time. The preprocessing process is to weight the pitch angle measurement value at the current collection time and the filter output value output after filtering by the first-order filter at the previous collection time to obtain the filter output value at the current collection time, and use the filter output value at each collection time as the preprocessing data; The calculation method of the first-order filter is: Among them, α represents the filter coefficient, X n Indicates the pitch angle measurement value at the current acquisition moment. Y (n-1) Indicates the filter output value at the last acquisition moment, Y n Indicates the filter output value at the current acquisition moment.

5. The method for detecting the retroreflection coefficient of a vehicle-mounted sign on an adaptive slope road according to claim 3, characterized in that: The specific process of step S23 is: S231, calculating the average value of n preprocessed data and the residual corresponding to each preprocessed data; S232, calculating the standard deviation of n preprocessed data according to the residual of each preprocessed data; S233, calculating the studentized residual corresponding to each preprocessed data according to the standard deviation and the residual of each preprocessed data; S234, traverse n pre-processed data, and determine whether the studentized residual corresponding to each pre-processed data exceeds a preset threshold. If so, remove the pre-processed data, and if not, save it in the road inclination angle data set; S235. Calculate the average value of the data in the road inclination data set to obtain the optimal estimated value of the road inclination.

6. A vehicle-mounted sign retroreflection coefficient detection system for adaptive slope roads, characterized in that: It includes an observation module, a detection module, a tilt sensor and an electric adjustment structure for adjusting the incident angle of the detection module, which are respectively connected to the control module for communication. The observation module includes a camera and a radar. The optical axes of the camera and the radar are installed on the roof of the inspection vehicle parallel to the length direction of the inspection vehicle to capture images of the area to be inspected. The detection module includes a detection camera, a light source for illumination, and a mounting bracket for fixing the detection camera and the light source, and the angle between the optical axis of the light source and the axis of the retroreflector of the traffic sign to be detected is adjusted to the target incident angle; The inclination sensor is installed horizontally on the top of the inspection vehicle to measure the road inclination information at any time when the inspection vehicle is driving on the road; The electric adjustment structure is installed on the luggage rack on the top of the inspection vehicle and is movably connected to the mounting bracket, and is used to drive the mounting bracket to move under the control of the control module so that the incident angle between the optical axis of the light source and the axis of the retroreflector of the traffic sign to be inspected remains unchanged; The control module is used to execute a method for detecting the retro-reflection coefficient of a vehicle-mounted sign on an adaptive slope road as described in any one of claims 1-5.

7. The vehicle-mounted sign retroreflection coefficient detection system for adaptive slope roads according to claim 6, characterized in that: The electric adjustment structure includes a stepper motor with a lead screw and an encoder. A lead screw guide rail is fixed on the top of the detection vehicle. A wedge-shaped slider is installed on the lead screw guide rail. The cross-section of the wedge-shaped slider is an isosceles right triangle with an acute angle of σ. The rotation of the lead screw drives the wedge-shaped slider to move back and forth along the lead screw guide rail. The wedge-shaped slider is connected to one end of the mounting bracket through a bearing, thereby raising or lowering the incident angle of the optical axis of the light source.

Images