Deformation threshold determination method and device, computer equipment and storage medium

Abstract

The invention relates to a deformation threshold determination method and device, computer equipment and a storage medium. The method comprises the following steps: acquiring initial deformation data collected by a sensor of a to-be-detected vehicle in a process that the to-be-detected vehicle runs on a reference road section; determining an initial deformation threshold value according to the dispersion degree of each initial deformation data; determining reference deformation data in the initial deformation data and predicted fluctuation positions corresponding to the reference deformation data according to the initial deformation threshold; for each reference deformation data, determining an actual fluctuation position corresponding to the reference deformation data according to the driving video data in the time period of the acquisition moment corresponding to the reference deformation data; and adjusting the initial deformation threshold value according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position. According to the technical scheme, the accuracy of the deformation threshold determination result is improved.

Description

Technical Field

[0001] The present application relates to the field of map technology, and in particular to a deformation threshold determination method, device, computer equipment and storage medium. Background Art

[0002] Vehicles may encounter uneven roads during driving, which may cause safety problems for people on board or affect the driving and riding experience, or even cause certain damage to the vehicle. Now, a deformation threshold is usually set to identify the unevenness of the road section. Therefore, the accuracy of the deformation threshold is crucial.

[0003] In conventional technology, historical deformation strengths of different road sections are usually statistically analyzed to obtain deformation thresholds of corresponding road sections. Although this method can complete the determination of deformation thresholds, it has the problem of low accuracy of the determined deformation thresholds. Summary of the invention

[0004] Based on this, it is necessary to provide a deformation threshold determination method, device, computer equipment and storage medium that can provide deformation threshold accuracy in response to the above technical problems.

[0005] In a first aspect, the present application provides a method for determining a deformation threshold, comprising:

[0006] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0007] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0008] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0009] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0010] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0011] In one of the embodiments, the actual fluctuation position corresponding to the reference deformation data is determined based on the driving video data within the time period corresponding to the collection moment of the reference deformation data, including: extracting the reference object in each video frame in the driving video data; determining whether the reference deformation data fluctuates at the collection moment according to the change of the posture of the reference object; if so, the collection position corresponding to the reference deformation data is used as the actual fluctuation position.

[0012] In one of the embodiments, based on the change in the posture of the reference object, it is determined whether the reference deformation data corresponding to the collection time fluctuates, including: determining the posture data of the reference object in each video frame; if the change amplitude of the posture data is greater than a preset amplitude threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates.

[0013] In one of the embodiments, if the change amplitude of the posture data is greater than a preset amplitude threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates, including: if the change amplitude of the angle data in the posture data is greater than the preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; or, if the change amplitude of the height data in the posture data is greater than the preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; wherein the angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

[0014] In one of the embodiments, the initial deformation threshold includes an upper deformation threshold and a lower deformation threshold; the upper deformation threshold is greater than the lower deformation threshold; accordingly, according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted, including: according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, determining the cause of the difference; wherein the cause of the difference is misidentification or missed identification; according to the cause of the difference, adjusting the upper deformation threshold and the lower deformation threshold.

[0015] In one of the embodiments, the upper deformation threshold and the lower deformation threshold are adjusted according to the cause of the difference, including: if the cause of the difference is misrecognition, the upper deformation threshold is increased, and the lower deformation threshold is lowered; if the cause of the difference is missed recognition, the upper deformation threshold is lowered, and the lower deformation threshold is increased.

[0016] In one of the embodiments, the upper deformation threshold is raised, and the lower deformation threshold is lowered, including: taking the maximum value of each misidentified reference deformation data that is greater than the upper deformation threshold as the adjusted upper deformation threshold, and taking the minimum value of each misidentified reference deformation data that is less than the lower deformation threshold as the adjusted lower deformation threshold.

[0017] In one of the embodiments, the upper deformation threshold is lowered and the lower deformation threshold is raised, including: if the cause of the difference is missed recognition, the maximum value of each missed reference deformation data that is less than the upper deformation threshold is used as the adjusted upper deformation threshold, and the minimum value of each missed reference deformation data that is greater than the lower deformation threshold is used as the adjusted lower deformation threshold.

[0018] In one of the embodiments, the initial deformation threshold is determined according to the degree of discreteness of each initial deformation data, including: grouping each initial deformation data according to the speed range to which the driving speed of the vehicle to be tested belongs; determining the initial deformation threshold in the corresponding speed range according to the degree of discreteness of the initial deformation data in the same group; accordingly, adjusting the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: adjusting the initial deformation threshold in the corresponding speed range according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0019] In a second aspect, the present application further provides a deformation threshold determination device, comprising:

[0020] A data acquisition module is used to acquire various initial deformation data collected by sensors of the vehicle under test when the vehicle under test is traveling on the reference road section;

[0021] A threshold determination module, used to determine the initial deformation threshold according to the discrete degree of each initial deformation data;

[0022] A position prediction module, used to determine each reference deformation data in each initial deformation data and a predicted fluctuation position corresponding to each reference deformation data according to an initial deformation threshold;

[0023] A position determination module is used to determine, for each reference deformation data, the actual fluctuation position corresponding to the reference deformation data according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0024] The threshold adjustment module is used to adjust the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0025] In a third aspect, the present application further provides a computer device, including a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the following steps are implemented:

[0026] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0027] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0028] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0029] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0030] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0031] In a fourth aspect, the present application further provides a computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the following steps are implemented:

[0032] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0033] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0034] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0035] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0036] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0037] In a fifth aspect, the present application further provides a computer program product, including a computer program, which implements the following steps when executed by a processor:

[0038] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0039] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0040] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0041] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0042] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0043] The above-mentioned deformation threshold determination method, device, computer equipment and storage medium obtain various initial deformation data collected by the sensor of the vehicle to be tested when the vehicle to be tested is driving on the reference road section; determine the initial deformation threshold according to the discrete degree of each initial deformation data; determine each reference deformation data in each initial deformation data and the predicted fluctuation position corresponding to each reference deformation data according to the initial deformation threshold; for each reference deformation data, determine the actual fluctuation position corresponding to the reference deformation data according to the driving video data in the time period corresponding to the collection time of the reference deformation data; adjust the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position. The above technical scheme introduces an initial deformation threshold to determine the reference deformation data and the predicted fluctuation position, and introduces the driving video data in the time period corresponding to the reference deformation data to determine the actual fluctuation position corresponding to the reference deformation data. According to the difference between the predicted fluctuation position and the actual fluctuation position corresponding to the reference deformation data, the initial deformation threshold determined based on the discrete degree of each initial deformation data is adjusted, so that the adjusted initial deformation threshold is more in line with the actual fluctuation of the road on the reference section, thereby improving the accuracy of the deformation threshold determination result. BRIEF DESCRIPTION OF THE DRAWINGS

[0044] In order to more clearly illustrate the technical solutions in the embodiments of the present application or the related technologies, the drawings required for use in the embodiments or the related technical descriptions are briefly introduced below. Obviously, the drawings described below are only some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying creative work.

[0045] Figure 1 A schematic diagram of a flow chart of a first deformation threshold determination method provided in this embodiment;

[0046] Figure 2 A schematic diagram of a flow chart of a step of determining an actual fluctuation position provided in this embodiment;

[0047] Figure 3 A schematic diagram of a flow chart of an initial deformation threshold adjustment step provided in this embodiment;

[0048] Figure 4A A schematic diagram of a flow chart of a step of determining an initial deformation threshold provided in this embodiment;

[0049] Figure 4B A schematic diagram of initial deformation data distribution provided in this embodiment;

[0050] Figure 5 A schematic diagram of a flow chart of a second deformation threshold determination method provided in this embodiment;

[0051] Figure 6 A structural block diagram of a deformation threshold determination device provided in this embodiment;

[0052] Figure 7 This is a diagram of the internal structure of a computer device provided in this embodiment. DETAILED DESCRIPTION

[0053] In order to make the purpose, technical solution and advantages of the present application more clearly understood, the present application is further described in detail below in conjunction with the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are only used to explain the present application and are not used to limit the present application.

[0054] In an exemplary embodiment, Figure 1 As shown, a method for determining a deformation threshold is provided, and the method is applied to Figure 1 The computer device in the example is used to illustrate, including the following steps S110 to S150. Among them:

[0055] S110, obtaining initial deformation data collected by sensors of the vehicle to be tested during the vehicle to be tested traveling on the reference road section.

[0056] The reference road section may be a road section that is at least partially fluctuating, that is, uneven. The deformation data may be data characterizing the degree of road fluctuation. The initial deformation data may be deformation data characterizing the degree of fluctuation of the reference road section, and the initial deformation data may be stored in a sensor log.

[0057] For example, the sensor log may include data such as the collection time, driving speed, fluctuation value, and collection location, and the present application does not impose any limitation on this. Accordingly, the initial deformation data may be generated according to the fluctuation value in the sensor log.

[0058] It should be noted that the initial deformation data may be the original deformation data collected by the sensor, or the data obtained after preprocessing the original deformation data. The preprocessing method may be to select a preset number (e.g., 10) of sensor logs of the vehicle to be tested within the same second, and use the extreme values ​​(including the maximum and minimum values) of the original deformation data in all sensor logs as the initial deformation data in the merged sensor logs.

[0059] S120: determining an initial deformation threshold according to the discreteness of each initial deformation data.

[0060] The initial deformation threshold may be a threshold used to roughly determine whether the road is undulating. The initial deformation threshold may include an upper deformation threshold and a lower deformation threshold; wherein the upper deformation threshold is greater than the lower deformation threshold.

[0061] Exemplarily, the mean value, variance value and standard deviation value of each initial deformation data are determined. For each initial deformation data, the absolute value corresponding to the first difference between the initial deformation data and the mean value is first determined, and then the second difference between the absolute value and the standard deviation value is determined; the initial deformation data corresponding to the second difference value greater than zero is used as discrete deformation data; the smallest discrete deformation data among all discrete deformation data greater than the mean value is selected as the upper deformation threshold, and the largest discrete deformation data among all discrete deformation data less than the mean value is selected as the lower deformation threshold; an initial deformation threshold including an upper deformation threshold and a lower deformation threshold is generated.

[0062] S130, determining each reference deformation data in each initial deformation data and a predicted fluctuation position corresponding to each reference deformation data according to the initial deformation threshold.

[0063] The reference deformation data may include initial deformation data greater than the deformation upper threshold value and initial deformation data less than the deformation lower threshold value among the initial deformation data. Alternatively, the reference deformation data may be initial deformation data adjacent to the deformation upper threshold value and initial deformation data close to the deformation lower threshold value among the initial deformation data. The predicted fluctuation position may be a collection position less than the deformation upper threshold value and a collection position corresponding to the initial deformation data greater than the deformation lower threshold value.

[0064] Exemplarily, from the initial deformation data, initial deformation data greater than the upper deformation threshold and initial deformation data adjacent to the lower deformation threshold are selected as reference deformation data; the sensor log corresponding to the reference deformation data is determined, and the collection position corresponding to the reference deformation data is extracted from the sensor log as the predicted fluctuation position.

[0065] S140, for each reference deformation data, according to the driving video data in the time period corresponding to the collection time of the reference deformation data, determine the actual fluctuation position corresponding to the reference deformation data.

[0066] The driving video may be a driving video of the vehicle collected by a driving recorder during the driving process of the vehicle. The actual fluctuation position may be obtained by performing image recognition analysis on the driving video data at the collection position corresponding to the reference deformation data.

[0067] In an optional embodiment, for each reference deformation data, driving video data within the time period corresponding to the collection moment of the reference deformation data is obtained, and a reference object is selected from each video frame of the driving video data to determine the posture change of the reference object within the time period corresponding to the collection moment of the reference deformation data; based on the posture change, it is determined whether the reference deformation data corresponding to the collection moment fluctuates, and the collection position corresponding to the reference deformation data when fluctuations occur is used as the actual fluctuation position.

[0068] Exemplarily, an optional method for obtaining driving video data in the time period corresponding to the acquisition time of the reference deformation data may be to determine the original driving video data in the time period corresponding to the acquisition time of the reference deformation data; for the first original video data acquired in the first preset time period (such as 2 seconds) before the acquisition time, and the second original video data acquired in the second preset time period (such as 3 seconds) after the acquisition time, the video data obtained by splicing the first original video data and the second original video data is used as the driving video data. Among them, the first preset time period may be the same as the second preset time period (such as both are 2 seconds), or they may be different, and this application does not impose any limitation on this.

[0069] S150, adjusting the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0070] Exemplarily, if the predicted fluctuation position represents the position point information of the specific existence of fluctuation situation predicted based on the initial deformation threshold, and the actual fluctuation position represents the position point information of the specific existence of fluctuation situation determined based on the driving video data, then the position point existing in the predicted fluctuation position but not in the actual fluctuation position can be regarded as the misidentified position point; and the position point existing in the actual fluctuation position but not in the predicted fluctuation position can be regarded as the missed identification position point; if the number of misidentified position points is greater than the number of missed identification position points, the upper deformation threshold is increased, and the lower deformation threshold is decreased. If the number of missed identification position points is greater than the number of misidentified position points, the upper deformation threshold is decreased, and the lower deformation threshold is increased.

[0071] Exemplarily, if the predicted fluctuation position characterizes whether there is fluctuation in each acquisition position predicted based on the initial deformation threshold, and the actual fluctuation position characterizes whether there is fluctuation in each acquisition position determined based on the driving video data, then the position points with fluctuation in the predicted fluctuation position and the position points without fluctuation in the actual fluctuation position can be intersected to obtain misidentified position points; and the position points with fluctuation in the predicted fluctuation position and the position points with fluctuation in the actual fluctuation position can be intersected to obtain missed identification position points; if the number of misidentified position points is greater than the number of missed identification position points, the upper deformation threshold is increased, and the lower deformation threshold is decreased. If the number of missed identification position points is greater than the number of misidentified position points, the upper deformation threshold is decreased, and the lower deformation threshold is increased.

[0072] The above-mentioned deformation threshold determination method obtains each initial deformation data collected by the sensor of the vehicle to be tested when the vehicle to be tested is driving on the reference road section; determines the initial deformation threshold according to the discrete degree of each initial deformation data; determines each reference deformation data in each initial deformation data and the predicted fluctuation position corresponding to each reference deformation data according to the initial deformation threshold; for each reference deformation data, determines the actual fluctuation position corresponding to the reference deformation data according to the driving video data in the time period corresponding to the collection time of the reference deformation data; adjusts the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position. The above technical scheme introduces an initial deformation threshold to determine the reference deformation data and the predicted fluctuation position, and introduces the driving video data in the time period corresponding to the reference deformation data to determine the actual fluctuation position corresponding to the reference deformation data. According to the difference between the predicted fluctuation position and the actual fluctuation position corresponding to the reference deformation data, the initial deformation threshold determined based on the discrete degree of each initial deformation data is adjusted, so that the adjusted initial deformation threshold is more in line with the actual fluctuation of the road on the reference section, thereby improving the accuracy of the deformation threshold determination result.

[0073] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the actual fluctuation position determination step in S140 is refined.

[0074] See also Figure 2 The steps for determining the actual fluctuation position shown include:

[0075] S210, extracting reference objects in each video frame of the driving video data.

[0076] Among them, the reference object can be an object in the driving video data that is easy to judge whether the road surface fluctuates, that is, an object that can be used to determine whether the vehicle to be tested fluctuates during driving. For example, since it is desired to determine whether the road surface fluctuates through video frames, it is necessary to compare the same reference object in each video frame, and the vehicle to be tested itself is relatively static relative to the driving recorder, and the dynamic part is only the real scene outside the car window, and the direction of fluctuation is horizontal in most cases. At the same time, the process of the vehicle to be tested driving on the reference road section is usually selected in the daytime, weather conditions and urban roads with good light, and the real scene outside the car window collected is relatively simple. Therefore, horizontal line objects can be selected as reference objects (such as corresponding dividing lines between the sky and the earth) to improve the accuracy of the actual fluctuation position determination result.

[0077] In one embodiment, extracting the reference object in each video frame in the driving video data includes: extracting each video frame in the driving video data; and extracting the reference object in the video frame by image analysis for each video frame.

[0078] Specifically, the method of extracting each video frame from the driving video data may be to extract a preset number (eg, 10 frames) of video frames from each second of the driving video data.

[0079] Accordingly, for each video frame, the method of extracting the reference object in the video frame may include: performing image preprocessing on the video frame to obtain the video frame after image preprocessing; extracting the first reference area and the second reference area in the video according to the color and / or texture in the video frame; identifying the edge lines of the first reference area and the second reference area through an edge detection algorithm, and taking the horizontal edge lines that overlap or are close to each other in the two areas as reference objects (i.e., relatively centered and long horizontal lines in the video frame). The first reference area may be adjacent to the second reference area, the first reference area may be a sky area, and the second reference area may be a ground area.

[0080] The image preprocessing may include at least one of denoising, contrast enhancement and color space conversion. Denoising is used to reduce the interference of noise in the video frame on image analysis. Contrast enhancement is used to make the color difference between the first reference area and the second reference area more obvious. Color space conversion is used to convert the video frame from the red, green, blue (RGB) color space to the hue, saturation, value (HSV) color space to better separate the color features of the first reference area and the second reference area.

[0081] In an optional embodiment, extracting a first reference area and a second reference area in a video according to the colors in a video frame may include: selecting an area with blue or white colors as the first reference area according to the color distribution in the video frame; and selecting an area containing at least three different colors as the second reference area.

[0082] In another optional embodiment, extracting a first reference area and a second reference area in a video based on a texture in a video frame may include: based on the texture smoothness in the video frame, selecting an area with a texture smoothness higher than a preset smoothness threshold as the first reference area; and selecting an area with a texture smoothness lower than a preset smoothness threshold as the second reference area.

[0083] In another optional embodiment, extracting a first reference area and a second reference area in a video based on the color and texture in the video frame may include: selecting an area with blue or white color and texture smoothness higher than a preset smoothness threshold as the first reference area based on the color distribution and texture smoothness in the video frame; selecting an area with more complex color and texture smoothness lower than a preset smoothness threshold as the second reference area.

[0084] S220, determining whether the reference deformation data fluctuates at the corresponding acquisition time according to the change of the posture of the reference object.

[0085] In one of the embodiments, based on the change in the posture of the reference object, it is determined whether the reference deformation data corresponding to the collection time fluctuates, including: determining the posture data of the reference object in each video frame; if the change amplitude of the posture data is greater than a preset amplitude threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates.

[0086] Specifically, the method for determining the posture data of the reference object in each video frame can be, for each video frame, determining the pixel height of the reference object in the video frame as the height data in the corresponding posture data of the reference object; and using the angle between the reference object and the lower edge of the video frame (i.e., the reference direction) as the angle data in the corresponding posture data of the reference object.

[0087] Specifically, if the change amplitude of the posture data is greater than the preset amplitude threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates, including: if the change amplitude of the angle data in the posture data is greater than the preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; or, if the change amplitude of the height data in the posture data is greater than the preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; wherein the angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

[0088] Exemplarily, if the change amplitude of the angle data in the posture data is greater than a preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates, including: for each video frame corresponding to each second of driving video data, determining the first absolute value of the difference between the angle data corresponding to two adjacent video frames in each video frame, and using the first absolute value as the change amplitude of the angle data; if there is a situation in which the change amplitude of the angle data in each video frame is greater than or equal to the preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates.

[0089] Exemplarily, if the change amplitude of the height data in the posture data is greater than a preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates, including: for each video frame corresponding to each second of driving video data, determining the second absolute value of the difference between the height data corresponding to two adjacent video frames in each video frame, and using the second absolute value as the change amplitude of the height data; if there is a situation in which the change amplitude of the height data in each video frame is greater than or equal to the preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates.

[0090] Exemplarily, for each video frame corresponding to each second of driving video data, if the change amplitude of the angle data in the posture data in each video frame is less than the preset angle threshold, and the change amplitude of the height data in the posture data is less than the preset height threshold, it is determined that the reference deformation data has not fluctuated at the corresponding acquisition time.

[0091] S230: If yes, the acquisition position corresponding to the reference deformation data is used as the actual fluctuation position.

[0092] Specifically, for the reference deformation data that fluctuates at the time of collection, the collection position corresponding to the reference deformation data is used as the actual fluctuation position.

[0093] The embodiment of the present application determines whether the reference deformation data fluctuates at the corresponding collection time by the change of the posture of the reference object in each video frame in the driving video data. If fluctuation occurs, the collection position corresponding to the reference deformation data is used as the actual fluctuation position, and the fluctuation position of the reference deformation data is determined by different methods, thereby improving the rationality of subsequent adjustment of the initial deformation threshold according to the difference between the actual fluctuation position and the predicted fluctuation position, thereby further making the adjusted initial deformation threshold more accurate.

[0094] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the initial deformation threshold adjustment step in S150 is refined.

[0095] See also Figure 3 The initial deformation threshold adjustment steps shown include:

[0096] S310, determining the cause of the difference based on the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0097] The reason for the difference may be misidentification or missed identification.

[0098] Specifically, for each reference deformation data, determine whether there is a fluctuation in the predicted fluctuation position corresponding to the reference deformation data, and whether there is a fluctuation in the actual fluctuation position corresponding to the reference deformation data; when there is a fluctuation in both the predicted fluctuation position and the actual fluctuation position, or when there is no fluctuation in both the predicted fluctuation position and the actual fluctuation position, it is proved that there is no difference between the predicted fluctuation position and the corresponding actual fluctuation position, and therefore, the reference deformation data is not misidentified or missed. If there is a fluctuation in the predicted fluctuation position, but there is no fluctuation in the actual fluctuation position, it indicates that the acquisition position where there is no fluctuation is misidentified as a position where there is fluctuation; if there is no fluctuation in the predicted fluctuation position, but there is fluctuation in the actual fluctuation position, it indicates that the acquisition position where there is fluctuation is missed as a position where there is no fluctuation; count the first number of misidentified acquisition positions, and the second number of missed acquisition positions; if the first number is greater than the second number, determine that the cause of the difference is misidentification; if the first number is less than the second number, determine that the cause of the difference is missed.

[0099] S320: adjusting the upper deformation threshold and the lower deformation threshold according to the difference reason.

[0100] In an optional embodiment, different upper bound adjustment amplitudes and lower bound adjustment amplitudes are pre-set for different difference reasons; according to the difference reason, the corresponding upper bound adjustment amplitude is selected to adjust the upper bound threshold of deformation, and the corresponding lower bound adjustment amplitude is selected to adjust the lower bound threshold of deformation. The upper bound adjustment amplitude and the lower bound adjustment amplitude can be the same or different. Exemplarily, if the difference reason is misidentification, the upper bound adjustment amplitude corresponding to the misidentification is +50, the lower bound adjustment amplitude is -60, the upper bound threshold of deformation is -1000, and the lower bound threshold of deformation is -1100, then, the upper bound threshold of deformation is adjusted to (-1000)+(+50)=-950 by the upper bound adjustment amplitude, and the adjusted upper bound threshold of deformation is -950; the lower bound threshold of deformation is adjusted to (-1100)+(-60)=-1160 by the lower bound adjustment amplitude, and the adjusted upper bound threshold of deformation is -1160.

[0101] In another optional embodiment, adjusting the upper deformation threshold and the lower deformation threshold according to the cause of the difference may include: if the cause of the difference is misrecognition, raising the upper deformation threshold and lowering the lower deformation threshold; if the cause of the difference is missed recognition, lowering the upper deformation threshold and raising the lower deformation threshold.

[0102] Specifically, the upper deformation threshold is raised, and the lower deformation threshold is lowered, including: taking the maximum value of each misidentified reference deformation data that is greater than the upper deformation threshold as the adjusted upper deformation threshold, and taking the minimum value of each misidentified reference deformation data that is less than the lower deformation threshold as the adjusted lower deformation threshold.

[0103] Specifically, the upper deformation threshold is lowered, and the lower deformation threshold is raised, including: if the reason for the difference is missed recognition, the maximum value of each reference deformation data that is less than the upper deformation threshold is used as the adjusted upper deformation threshold, and the minimum value of each reference deformation data that is greater than the lower deformation threshold is used as the adjusted lower deformation threshold.

[0104] The embodiment of the present application determines the cause of the difference based on the difference between each predicted fluctuation position and the corresponding actual fluctuation position, and adjusts the upper deformation threshold and the lower deformation threshold according to the cause of the difference, so as to avoid the influence of misidentification and missed identification problems on road fluctuation identification, and further improve the accuracy and rationality of the determined initial deformation threshold.

[0105] Based on the technical solutions of the above embodiments, the present application also provides an optional embodiment. In this optional embodiment, the initial deformation threshold determination step in S120 is refined.

[0106] See also Figure 4A The steps for determining the initial deformation threshold shown include:

[0107] S410, grouping the initial deformation data according to the speed interval to which the driving speed of the vehicle to be tested belongs.

[0108] In one of the embodiments, each initial deformation data is grouped according to the speed range to which the driving speed of the vehicle to be tested belongs, including: merging each initial deformation data collected within the same collection time period (such as every second) of the vehicle to be tested, and merging the driving speeds corresponding to each initial deformation data; dividing at least one speed range according to each merged driving speed; and grouping each initial deformation data according to each speed range.

[0109] Specifically, merging the initial deformation data and merging the driving speeds corresponding to the initial deformation data may include: selecting a preset number (such as 10) of initial deformation data collected by the vehicle to be tested within the same collection period, and the driving speeds corresponding to the initial deformation data; selecting extreme values ​​(including maximum values ​​and minimum values) from the initial deformation data as the merged initial deformation data; determining the average value of the driving speeds corresponding to the initial deformation data, and using the average value as the merged driving speed.

[0110] Exemplarily, dividing at least one speed interval according to the merged driving speeds may include: setting at least one speed interval according to the merged driving speeds of the vehicle to be tested in each collection period, such as setting 0km / h-20km / h as the first speed interval, 20km / h-40km / h as the second speed interval, 40km / h-65km / h as the first speed interval... and so on.

[0111] Exemplarily, the initial deformation data may be grouped according to each speed interval by selecting, for each speed interval, a sensor log corresponding to the vehicle speed in the speed interval, and using the initial deformation data in the sensor log as the initial deformation data in the speed interval.

[0112] S420: determining an initial deformation threshold in a corresponding speed range according to the discreteness of the initial deformation data in the same group.

[0113] In one of the embodiments, determining the initial deformation threshold value in the corresponding speed interval according to the degree of discreteness of the initial deformation data in the same group may include: determining, for each speed interval, the average value of the initial deformation data in the speed interval; determining the variance and standard deviation of the initial deformation data in the speed interval according to the initial deformation data and the average value; determining the degree of discreteness of the initial deformation data in the speed interval according to the average value and the standard deviation; determining the initial deformation threshold value in the speed interval according to the degree of discreteness in the speed interval.

[0114] Exemplarily, the sum of all initial deformation data in the speed interval can be determined, and the quotient between the sum and the total number of initial deformation data can be used as the average value of the initial deformation data in the speed interval. For each initial deformation data in the speed interval, the square result of the absolute value between the initial deformation data and the average value is determined, and the sum of all the square results is used as the variance of the initial deformation data in the speed interval. The quotient between the variance and the total number of initial deformation data is determined, and the square root of the quotient is used as the standard deviation of the initial deformation data in the speed interval.

[0115] Exemplarily, determining the degree of discreteness of the initial deformation data in the speed range based on the mean value, variance value and standard deviation value can include: determining the absolute value corresponding to the first difference between the initial deformation data and the mean value; determining the second difference between the absolute value and the standard deviation value; if the second difference is greater than zero, it is considered that the initial deformation data is discrete.

[0116] Exemplarily, discrete initial deformation data may be selected, and the minimum value of the initial deformation data greater than the average value may be used as the upper deformation threshold; and the maximum value of the initial deformation data less than the average value may be used as the lower deformation threshold.

[0117] For example, Figure 4B The schematic diagram of the distribution of initial deformation data is shown, in which the ordinate represents the fluctuation value corresponding to the initial deformation data, the blue area represents all the initial deformation data greater than the average value, and the orange area represents all the initial deformation data less than the average value. In this embodiment, the smallest initial deformation data can be selected from all the initial deformation data greater than the average value corresponding to the blue area as the upper deformation threshold; and the largest initial deformation data can be selected from all the initial deformation data less than the average value in the orange area as the lower deformation threshold.

[0118] It should be noted that the initial deformation threshold is adjusted according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold in the corresponding speed range is adjusted.

[0119] Specifically, for each speed range, the predicted fluctuation position corresponding to each reference deformation data in the speed range and the actual fluctuation position corresponding to the reference deformation data are determined; for the predicted fluctuation position corresponding to each reference deformation data in each speed range, the difference between the predicted fluctuation position and the corresponding actual fluctuation position is determined, and the initial deformation threshold in the corresponding speed range is adjusted according to the difference.

[0120] Since the initial deformation data fed back by the sensor are different when the vehicle to be tested passes the same road surface fluctuation position at different driving speeds, the embodiment of the present application improves the accuracy of determining the initial deformation threshold by dividing the driving speed of the vehicle to be tested into speed intervals and determining the initial deformation threshold in the corresponding speed interval.

[0121] In one embodiment, this embodiment provides an optional method for determining a deformation threshold, and takes the method applied to a computer device as an example for illustration. Figure 5 As shown, the method comprises the following steps:

[0122] S510, obtaining initial deformation data collected by sensors of the vehicle to be tested during the vehicle to be tested traveling on the reference road section.

[0123] S520, grouping the initial deformation data according to the speed interval to which the driving speed of the vehicle to be tested belongs.

[0124] S530: determining an initial deformation threshold in a corresponding speed range according to the discreteness of the initial deformation data in the same group.

[0125] S540: Determine, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data.

[0126] S550: extracting the reference object in each video frame of the driving video data for each reference deformation data, and determining the position and posture data of the reference object in each video frame.

[0127] S560: If the change amplitude of the posture data is greater than the preset amplitude threshold, it is determined that the reference deformation data fluctuates at the corresponding collection time, and the collection position corresponding to the reference deformation data is used as the actual fluctuation position.

[0128] It should be noted that if the change amplitude of the angle data in the posture data is greater than the preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; or, if the change amplitude of the height data in the posture data is greater than the preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; wherein, the angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

[0129] S570, determining the cause of the difference in the corresponding speed range according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0130] Among them, the reasons for the difference are misidentification or missed identification.

[0131] S580: Adjust the initial deformation threshold in the corresponding speed range according to the difference reason in the corresponding speed range.

[0132] The initial deformation threshold includes an upper deformation threshold and a lower deformation threshold; the upper deformation threshold is greater than the lower deformation threshold.

[0133] It should be noted that if the reason for the difference is misidentification, the maximum value of each misidentified reference deformation data that is greater than the upper deformation threshold will be used as the adjusted upper deformation threshold, and the minimum value of each misidentified reference deformation data that is less than the lower deformation threshold will be used as the adjusted lower deformation threshold; if the reason for the difference is missed identification, the maximum value of each missed identification reference deformation data that is less than the upper deformation threshold will be used as the adjusted upper deformation threshold, and the minimum value of each missed identification reference deformation data that is greater than the lower deformation threshold will be used as the adjusted lower deformation threshold.

[0134] It should be noted that this embodiment can also develop a data point upload function on the vehicle-mounted end of the target vehicle: periodically collect the collection time, driving speed, deformation data and collection location in the sensor log of the target vehicle; through the initial deformation threshold corresponding to the speed interval to which the driving speed belongs, filter out the sensor log corresponding to the fluctuation data (i.e., deformation data greater than the upper deformation threshold, or deformation data less than the lower deformation threshold), and upload the sensor log to the cloud. After receiving the sensor log uploaded by the vehicle-mounted end, the cloud performs data cleaning, data conversion and persistent storage on the sensor log.

[0135] It should be noted that in order to save the data transmission traffic of the target vehicle and reduce the number of interactions between the vehicle-mounted terminal of the target vehicle and the cloud, so as to improve the operating performance of the vehicle-mounted terminal and the cloud, this embodiment can also detect whether there is continuous fluctuation data within a preset period (such as 1 minute and 30 seconds), and generate an upload request based on the fluctuation data within the preset period, and upload the sensor log corresponding to the fluctuation data to the cloud. This embodiment can also reduce the number of interactions between the vehicle-mounted terminal of the target vehicle and the cloud by controlling the maximum amount of data per time, or controlling the maximum amount of data uploaded daily by the vehicle-mounted terminal corresponding to the same target vehicle.

[0136] It should be noted that the present embodiment may also set up a "fluctuation layer production service", that is, according to the fluctuation position in the sensor log uploaded by the vehicle-mounted end (that is, the collection position of the deformation data in the sensor log), the fluctuation position is mounted on the road network data that provides the navigation function. Exemplarily, according to the fluctuation position in the sensor log uploaded by the vehicle-mounted end, the fluctuation position is mounted on the road network data that provides the navigation function, which may include: using the collection position in the sensor log as the fluctuation position; obtaining the longitude and latitude of the fluctuation point and the link number (Link ID) corresponding to the fluctuation position, and searching the longitude and latitude of the starting point of the road corresponding to the link number in the road network data; determining the distance between the starting point and the fluctuation point according to the longitude and latitude of the starting point and the longitude and latitude of the fluctuation point; and mounting the fluctuation position on the road corresponding to the link number in the road network data according to the distance between the starting point and the fluctuation point.

[0137] It should be noted that, in the process of providing route calculation guidance for the vehicle through the road network data, this embodiment can provide "avoidance detour" guidance or advance voice prompts to slow down and other processing for the fluctuating position of the route. This embodiment can also provide "avoidance detour" guidance or advance voice prompts to slow down and other processing for the fluctuating position of the route based on the fluctuating position mounted on the road network data in the autonomous driving scenario.

[0138] It should be understood that, although the various steps in the flowcharts involved in the above-mentioned embodiments are displayed in sequence according to the indication of the arrows, these steps are not necessarily executed in sequence according to the order indicated by the arrows. Unless there is a clear explanation in this article, the execution of these steps does not have a strict order restriction, and these steps can be executed in other orders. Moreover, at least a part of the steps in the flowcharts involved in the above-mentioned embodiments can include multiple steps or multiple stages, and these steps or stages are not necessarily executed at the same time, but can be executed at different times, and the execution order of these steps or stages is not necessarily to be carried out in sequence, but can be executed in turn or alternately with other steps or at least a part of the steps or stages in other steps.

[0139] Based on the same inventive concept, the embodiment of the present application also provides a deformation threshold determination device for implementing the deformation threshold determination method involved above. The implementation scheme for solving the problem provided by the device is similar to the implementation scheme recorded in the above method, so the specific limitations in the one or more deformation threshold determination device embodiments provided below can refer to the limitations of the deformation threshold determination method above, and will not be repeated here.

[0140] In an exemplary embodiment, Figure 6 As shown, a deformation threshold determination device is provided, including: a data acquisition module 610, a threshold determination module 620, a position prediction module 630, a position determination module 640 and a threshold adjustment module 650, wherein:

[0141] The data acquisition module 610 is used to acquire various initial deformation data collected by the sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0142] A threshold determination module 620, used to determine an initial deformation threshold according to the discrete degree of each initial deformation data;

[0143] The position prediction module 630 is used to determine each reference deformation data in each initial deformation data and a predicted fluctuation position corresponding to each reference deformation data according to the initial deformation threshold value;

[0144] The position determination module 640 is used to determine the actual fluctuation position corresponding to each reference deformation data according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0145] The threshold adjustment module 650 is used to adjust the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0146] In one embodiment, Figure 6 The location determination module 630 in the embodiment further comprises:

[0147] A reference extraction unit, used to extract reference objects in each video frame in the driving video data;

[0148] A fluctuation determination unit, used to determine whether the reference deformation data fluctuates at a corresponding acquisition time according to the change of the posture of the reference object;

[0149] The position determination unit is used to, if yes, use the acquisition position corresponding to the reference deformation data as the actual fluctuation position.

[0150] In one embodiment, the fluctuation determination unit further includes:

[0151] A posture determination subunit, used for the posture data of the reference object in each video frame;

[0152] The fluctuation determination subunit is used to determine that the reference deformation data fluctuates at the corresponding acquisition time if the change amplitude of the posture data is greater than a preset amplitude threshold.

[0153] In one of the embodiments, the fluctuation determination subunit is also used to determine that the reference deformation data fluctuates at the corresponding collection moment if the change amplitude of the angle data in the posture data is greater than a preset angle threshold; or, if the change amplitude of the height data in the posture data is greater than a preset height threshold, then determine that the reference deformation data fluctuates at the corresponding collection moment; wherein the angle data is the angle between the reference line of the reference object and the reference direction; and the height data is the pixel height of the reference line of the reference object.

[0154] In one embodiment, Figure 6 The threshold adjustment module 650 further includes:

[0155] A cause determination unit, for determining the cause of the difference according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position; wherein the cause of the difference is misidentification or missed identification;

[0156] The threshold adjustment unit is used to adjust the upper deformation threshold and the lower deformation threshold according to the difference reason.

[0157] In one embodiment, the threshold adjustment unit further includes:

[0158] A misidentification subunit, used to increase the upper threshold of deformation and decrease the lower threshold of deformation if the cause of the difference is misidentification;

[0159] The missed recognition subunit is used to lower the upper threshold of deformation and raise the lower threshold of deformation if the cause of the difference is missed recognition.

[0160] In one of the embodiments, the misidentification sub-unit is further used to use the maximum value of each misidentified reference deformation data that is greater than the deformation upper threshold as the adjusted deformation upper threshold, and to use the minimum value of each misidentified reference deformation data that is less than the deformation lower threshold as the adjusted deformation lower threshold.

[0161] In one of the embodiments, the missed identification sub-unit is also used to, if the cause of the difference is missed identification, use the maximum value of each reference deformation data that is less than the upper deformation threshold as the adjusted upper deformation threshold, and use the minimum value of each reference deformation data that is greater than the lower deformation threshold as the adjusted lower deformation threshold.

[0162] In one embodiment, Figure 6 The threshold determination module 620 in the method is also used to group the initial deformation data according to the speed range to which the driving speed of the vehicle to be tested belongs; determine the initial deformation threshold value in the corresponding speed range according to the degree of discreteness of the initial deformation data in the same group; accordingly, adjust the initial deformation threshold value according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: adjusting the initial deformation threshold value in the corresponding speed range according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0163] Each module in the above deformation threshold determination device can be implemented in whole or in part by software, hardware, or a combination thereof. Each module can be embedded in or independent of a processor in a computer device in the form of hardware, or can be stored in a memory in a computer device in the form of software, so that the processor can call and execute operations corresponding to each module.

[0164] In an exemplary embodiment, a computer device is provided. The computer device may be a terminal, and its internal structure diagram may be as shown in FIG. Figure 7As shown. The computer device includes a processor, a memory, an input / output interface, a communication interface, a display unit and an input device. The processor, the memory and the input / output interface are connected through a system bus, and the communication interface, the display unit and the input device are connected to the system bus through the input / output interface. The processor of the computer device is used to provide computing and control capabilities. The memory of the computer device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The input / output interface of the computer device is used to exchange information between the processor and the external device. The communication interface of the computer device is used to communicate with an external terminal in a wired or wireless manner, and the wireless manner can be realized through WIFI, a mobile cellular network, NFC (near field communication) or other technologies. When the computer program is executed by the processor, a deformation threshold determination method is implemented. The display unit of the computer device is used to form a visually visible picture, which can be a display screen, a projection device or a virtual reality imaging device. The display screen can be a liquid crystal display screen or an electronic ink display screen, and the input device of the computer device can be a touch layer covering the display screen, or a button, trackball or touchpad set on the computer device shell, or an external keyboard, touchpad or mouse.

[0165] Those skilled in the art will understand that Figure 7 The structure shown in the figure is only a block diagram of a part of the structure related to the solution of the present application, and does not constitute a limitation on the computer device to which the solution of the present application is applied. The specific computer device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.

[0166] In an exemplary embodiment, a computer device is provided, including a memory and a processor, wherein a computer program is stored in the memory, and when the processor executes the computer program, the following steps are implemented:

[0167] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0168] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0169] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0170] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0171] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0172] In one embodiment, when the processor executes the computer program, the following steps are also implemented: extracting the reference object in each video frame in the driving video data; determining whether the reference deformation data fluctuates at the corresponding acquisition moment according to the change in the posture of the reference object; if so, taking the acquisition position corresponding to the reference deformation data as the actual fluctuation position.

[0173] In one embodiment, when the processor executes the computer program, the following steps are also implemented: determining the posture data of the reference object in each video frame; if the change amplitude of the posture data is greater than a preset amplitude threshold, determining that the reference deformation data fluctuates at the corresponding acquisition time.

[0174] In one embodiment, when the processor executes the computer program, the following steps are also implemented: if the change amplitude of the angle data in the posture data is greater than a preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; or, if the change amplitude of the height data in the posture data is greater than a preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; wherein the angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

[0175] In one embodiment, when the processor executes the computer program, the following steps are also implemented: according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the cause of the difference is determined; wherein the cause of the difference is misidentification or missed identification; according to the cause of the difference, the upper deformation threshold and the lower deformation threshold are adjusted.

[0176] In one embodiment, when the processor executes the computer program, the following steps are also implemented: if the cause of the difference is misrecognition, the upper deformation threshold is increased, and the lower deformation threshold is decreased; if the cause of the difference is missed recognition, the upper deformation threshold is decreased, and the lower deformation threshold is increased.

[0177] In one embodiment, when the processor executes the computer program, the following steps are also implemented: the maximum value of each incorrectly identified reference deformation data that is greater than the deformation upper threshold is used as the adjusted deformation upper threshold, and the minimum value of each incorrectly identified reference deformation data that is less than the deformation lower threshold is used as the adjusted deformation lower threshold.

[0178] In one embodiment, when the processor executes the computer program, the following steps are also implemented: if the cause of the difference is missed recognition, the maximum value of each reference deformation data that is less than the upper deformation threshold is used as the adjusted upper deformation threshold, and the minimum value of each reference deformation data that is greater than the lower deformation threshold is used as the adjusted lower deformation threshold.

[0179] In one embodiment, the processor also implements the following steps when executing the computer program: grouping each initial deformation data according to the speed range to which the driving speed of the vehicle to be tested belongs; determining the initial deformation threshold in the corresponding speed range according to the degree of discreteness of the initial deformation data in the same group; accordingly, adjusting the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: adjusting the initial deformation threshold in the corresponding speed range according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0180] In one embodiment, a computer readable storage medium is provided, on which a computer program is stored, and when the computer program is executed by a processor, the following steps are implemented:

[0181] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0182] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0183] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0184] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0185] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0186] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: extracting the reference object in each video frame in the driving video data; determining whether the reference deformation data fluctuates at the corresponding acquisition moment based on the change in the posture of the reference object; if so, taking the acquisition position corresponding to the reference deformation data as the actual fluctuation position.

[0187] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the posture data of the reference object in each video frame; if the change amplitude of the posture data is greater than a preset amplitude threshold, determining that the reference deformation data fluctuates at the corresponding acquisition time.

[0188] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the change amplitude of the angle data in the posture data is greater than a preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; or, if the change amplitude of the height data in the posture data is greater than a preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; wherein the angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

[0189] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the cause of the difference is determined; wherein the cause of the difference is misidentification or missed identification; according to the cause of the difference, the upper deformation threshold and the lower deformation threshold are adjusted.

[0190] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the cause of the difference is misrecognition, the upper deformation threshold is increased, and the lower deformation threshold is decreased; if the cause of the difference is missed recognition, the upper deformation threshold is decreased, and the lower deformation threshold is increased.

[0191] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented: the maximum value of each incorrectly identified reference deformation data that is greater than the deformation upper threshold is used as the adjusted deformation upper threshold, and the minimum value of each incorrectly identified reference deformation data that is less than the deformation lower threshold is used as the adjusted deformation lower threshold.

[0192] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the cause of the difference is missed recognition, the maximum value of each reference deformation data that is less than the upper deformation threshold is used as the adjusted upper deformation threshold, and the minimum value of each reference deformation data that is greater than the lower deformation threshold is used as the adjusted lower deformation threshold.

[0193] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: grouping the initial deformation data according to the speed range to which the driving speed of the vehicle to be tested belongs; determining the initial deformation threshold in the corresponding speed range according to the degree of discreteness of the initial deformation data in the same group; accordingly, adjusting the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: adjusting the initial deformation threshold in the corresponding speed range according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0194] In one embodiment, a computer program product is provided, comprising a computer program, which, when executed by a processor, implements the following steps:

[0195] Acquire various initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section;

[0196] According to the discrete degree of each initial deformation data, the initial deformation threshold is determined;

[0197] Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data;

[0198] For each reference deformation data, the actual fluctuation position corresponding to the reference deformation data is determined according to the driving video data in the time period corresponding to the collection time of the reference deformation data;

[0199] According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold is adjusted.

[0200] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: extracting the reference object in each video frame in the driving video data; determining whether the reference deformation data fluctuates at the corresponding acquisition moment based on the change in the posture of the reference object; if so, taking the acquisition position corresponding to the reference deformation data as the actual fluctuation position.

[0201] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: determining the posture data of the reference object in each video frame; if the change amplitude of the posture data is greater than a preset amplitude threshold, determining that the reference deformation data fluctuates at the corresponding acquisition time.

[0202] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the change amplitude of the angle data in the posture data is greater than a preset angle threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; or, if the change amplitude of the height data in the posture data is greater than a preset height threshold, it is determined that the reference deformation data corresponding to the collection time fluctuates; wherein the angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

[0203] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the cause of the difference is determined; wherein the cause of the difference is misidentification or missed identification; according to the cause of the difference, the upper deformation threshold and the lower deformation threshold are adjusted.

[0204] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the cause of the difference is misrecognition, the upper deformation threshold is increased, and the lower deformation threshold is decreased; if the cause of the difference is missed recognition, the upper deformation threshold is decreased, and the lower deformation threshold is increased.

[0205] In one embodiment, when the computer program is executed by a processor, the following steps are also implemented: the maximum value of each incorrectly identified reference deformation data that is greater than the deformation upper threshold is used as the adjusted deformation upper threshold, and the minimum value of each incorrectly identified reference deformation data that is less than the deformation lower threshold is used as the adjusted deformation lower threshold.

[0206] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: if the cause of the difference is missed recognition, the maximum value of each reference deformation data that is less than the upper deformation threshold is used as the adjusted upper deformation threshold, and the minimum value of each reference deformation data that is greater than the lower deformation threshold is used as the adjusted lower deformation threshold.

[0207] In one embodiment, when the computer program is executed by the processor, the following steps are also implemented: grouping the initial deformation data according to the speed range to which the driving speed of the vehicle to be tested belongs; determining the initial deformation threshold in the corresponding speed range according to the degree of discreteness of the initial deformation data in the same group; accordingly, adjusting the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: adjusting the initial deformation threshold in the corresponding speed range according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

[0208] It should be noted that the data involved in this application (including but not limited to data used for analysis, stored data, displayed data, etc.) are all information and data authorized by the user or fully authorized by all parties, and the collection, use and processing of relevant data must comply with relevant regulations.

[0209] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to the memory, database or other medium used in the embodiments provided in this application can include at least one of non-volatile and volatile memory. Non-volatile memory can include read-only memory (ROM), magnetic tape, floppy disk, flash memory, optical memory, high-density embedded non-volatile memory, resistive random access memory (ReRAM), magnetoresistive random access memory (MRAM), ferroelectric random access memory (FRAM), phase change memory (PCM), graphene memory, etc. Volatile memory can include random access memory (RAM) or external cache memory, etc. As an illustration and not limitation, RAM can be in various forms, such as static random access memory (SRAM) or dynamic random access memory (DRAM). The database involved in each embodiment provided in this application may include at least one of a relational database and a non-relational database. Non-relational databases may include distributed databases based on blockchains, etc., but are not limited to this. The processor involved in each embodiment provided in this application may be a general-purpose processor, a central processing unit, a graphics processor, a digital signal processor, a programmable logic device, a data processing logic device based on quantum computing, etc., but are not limited to this.

[0210] The technical features of the above embodiments may be combined arbitrarily. To make the description concise, not all possible combinations of the technical features in the above embodiments are described. However, as long as there is no contradiction in the combination of these technical features, they should be considered to be within the scope of this specification.

[0211] The above-described embodiments only express several implementation methods of the present application, and the descriptions thereof are relatively specific and detailed, but they cannot be understood as limiting the scope of the present application. It should be pointed out that, for a person of ordinary skill in the art, several variations and improvements can be made without departing from the concept of the present application, and these all belong to the protection scope of the present application. Therefore, the protection scope of the present application shall be subject to the attached claims.

Claims

1. A method for determining a deformation threshold, characterized in that: include: Acquiring initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section; Determining an initial deformation threshold according to the discrete degree of each of the initial deformation data; Determining, according to the initial deformation threshold, each reference deformation data in each initial deformation data, and a predicted fluctuation position corresponding to each reference deformation data; For each of the reference deformation data, according to the driving video data in the time period corresponding to the collection time of the reference deformation data, determine the actual fluctuation position corresponding to the reference deformation data; The initial deformation threshold is adjusted according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

2. The method according to claim 1, characterized in that The determining, based on the driving video data in the time period corresponding to the collection time of the reference deformation data, the actual fluctuation position corresponding to the reference deformation data comprises: Extracting reference objects in each video frame of the driving video data; Determining whether the reference deformation data fluctuates at a corresponding acquisition time according to a change in the posture of the reference object; If so, the acquisition position corresponding to the reference deformation data is used as the actual fluctuation position.

3. The method according to claim 2, characterized in that The step of determining whether the reference deformation data fluctuates at a corresponding acquisition time according to the change in the position and posture of the reference object includes: Determining the position and posture data of the reference object in each of the video frames; If the change amplitude of the posture data is greater than a preset amplitude threshold, it is determined that the reference deformation data fluctuates at the corresponding acquisition time.

4. The method according to claim 3, characterized in that If the change amplitude of the posture data is greater than a preset amplitude threshold, determining that the reference deformation data fluctuates at the corresponding acquisition time includes: If the change range of the angle data in the posture data is greater than the preset angle threshold, it is determined that the reference deformation data fluctuates at the corresponding acquisition time; or, If the change amplitude of the height data in the posture data is greater than a preset height threshold, it is determined that the reference deformation data fluctuates at the corresponding acquisition time; The angle data is the angle between the reference line of the reference object and the reference direction; the height data is the pixel height of the reference line of the reference object.

5. The method according to claim 1, characterized in that: The initial deformation threshold includes an upper deformation threshold and a lower deformation threshold; the upper deformation threshold is greater than the lower deformation threshold; accordingly, the initial deformation threshold is adjusted according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: Determining the cause of the difference according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position; wherein the cause of the difference is misidentification or missed identification; According to the cause of the difference, the upper deformation threshold and the lower deformation threshold are adjusted.

6. The method according to claim 5, characterized in that The adjusting the deformation upper threshold and the deformation lower threshold according to the difference reason includes: If the cause of the difference is misidentification, the upper threshold of deformation is increased, and the lower threshold of deformation is decreased; If the cause of the difference is missed recognition, the upper deformation threshold is lowered, and the lower deformation threshold is increased.

7. The method according to claim 6, characterized in that The step of increasing the upper threshold of deformation and decreasing the lower threshold of deformation includes: The maximum value of each incorrectly identified reference deformation data that is greater than the deformation upper threshold is used as the adjusted deformation upper threshold, and the minimum value of each incorrectly identified reference deformation data that is less than the deformation lower threshold is used as the adjusted deformation lower threshold.

8. The method according to claim 6, characterized in that The lowering of the deformation upper threshold and the raising of the deformation lower threshold include: If the cause of the difference is missed recognition, the maximum value of each missed reference deformation data that is less than the upper deformation threshold is used as the adjusted upper deformation threshold, and the minimum value of each missed reference deformation data that is greater than the lower deformation threshold is used as the adjusted lower deformation threshold.

9. The method according to claim 1, characterized in that: Determining the initial deformation threshold according to the discrete degree of each of the initial deformation data includes: Grouping the initial deformation data according to the speed interval to which the driving speed of the vehicle to be tested belongs; According to the discrete degree of the initial deformation data in the same group, the initial deformation threshold value in the corresponding speed range is determined; Accordingly, the initial deformation threshold is adjusted according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, including: According to the difference between each predicted fluctuation position and the corresponding actual fluctuation position, the initial deformation threshold in the corresponding speed range is adjusted.

10. A deformation threshold determination device, characterized in that: include: A data acquisition module, used to acquire initial deformation data collected by sensors of the vehicle under test during the vehicle under test driving on the reference road section; A threshold determination module, used to determine an initial deformation threshold according to the discrete degree of each of the initial deformation data; A position prediction module, used to determine each reference deformation data in each of the initial deformation data and a predicted fluctuation position corresponding to each of the reference deformation data according to the initial deformation threshold; A position determination module, for determining, for each of the reference deformation data, an actual fluctuation position corresponding to the reference deformation data according to the driving video data in the time period corresponding to the collection time of the reference deformation data; The threshold adjustment module is used to adjust the initial deformation threshold according to the difference between each predicted fluctuation position and the corresponding actual fluctuation position.

11. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

12. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 9 are implemented.

Images