Test method, device and system

Abstract

A test method, device and system relate to the technical field of vehicle testing, the method obtains a first feature of virtual image information and a second feature of live-action information based on a first image by obtaining the first image, the first feature being used as an attribute representation of the virtual image information, the second feature being used as an attribute representation of the live-action information, and the second feature being used as an attribute representation of the live-action information; the attribute comprises a position and / or a direction; and determining a fitting degree evaluation result of the first HUD based on the first feature and the second feature, so that the AR effect of the HUD can be accurately evaluated.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of vehicle testing, and in particular to a testing method, device and system. BACKGROUND

[0002] AR-HUD is a technology that combines augmented reality (AR) with head-up display (HUD). Augmented reality refers to the application of virtual information generated by a computer to the real world through various technical means such as multimedia, three-dimensional modeling, real-time tracking, intelligent interaction and sensing, so as to enhance the real world. Head-up display refers to displaying auxiliary driving information at a position that can be seen by the driver looking up (looking straight), so that the driver can know information such as cruise guidance information, instrument data and other driving-related information without looking down at the instrument or other information display device.

[0003] Currently, the industry mostly relies on subjective judgment of videos observed by driving vehicles for testing of AR effects. This method not only consumes time and effort, but is also easily affected by testers' personal experience and subjective feelings, resulting in evaluation results that cannot objectively evaluate the ability and effect of the product.

[0004] In summary, the AR effect evaluation method of the existing HUD has the problems of strong subjectivity, lack of unified standard and low testing efficiency. SUMMARY

[0005] The present application provides a testing method, device and system, which can accurately evaluate the AR effect of the HUD.

[0006] The testing method provided by the embodiments of the present application can be executed by a computing device or a module in the computing device. The computing device refers to a device that can be abstracted as a computer system, wherein the computing device supporting the testing function can also be referred to as a testing device. For ease of description, the following will be described by taking the execution subject as the testing device. It should be understood that the testing device can be a whole machine, such as a smart phone, a tablet computer, a notebook computer, a desktop computer, a car machine, a vehicle-mounted computer or a server, etc., or a vehicle-mounted system / device composed of multiple whole machines, or the testing device can also include software and / or hardware modules that do not constitute a whole machine, such as a testing function related chip, such as a system chip, wherein the system chip is also referred to as a system on chip, or a SoC chip. As a possible example, the computing device / testing device can be a terminal device such as a car machine or a vehicle-mounted computer in a smart vehicle, or a system chip or a testing chip that can be arranged in a computer system or a surround view system of a smart terminal.

[0007] In a first aspect, a testing method is provided. The method comprises: obtaining, by a testing device, a first image, the first image being an image from a perspective of a driver of a test vehicle, the test vehicle comprising a first HUD, the first HUD being configured to generate a projected image, a position of the projected image being located in the perspective of the driver of the test vehicle. The first image comprises real scene information and virtual image information, the real scene information being elements in a real driving environment of the test vehicle, and the virtual image information being elements in the projected image that are related to the real scene information.

[0008] The testing device obtains, based on the first image, a first feature of the virtual image information and a second feature of the real scene information. The first feature is used to represent an attribute of the virtual image information, and the second feature is used to represent an attribute of the real scene information. The attribute includes a position and / or a direction. The virtual image information comprises cruise control indication information and / or attribute indication information. The attribute indication information comprises one or more of a pedestrian identifier, an obstacle identifier, a front vehicle identifier, or a lane line identifier.

[0009] The testing device determines, based on the first feature and the second feature, a fitting degree evaluation result of the first HUD. The fitting degree evaluation result is used to indicate a fitting degree of the projected image of the first HUD with the real driving environment.

[0010] With the method provided in the first aspect, the testing device can obtain an image from the perspective of the driver of the test vehicle, so that the image contains real scene information and virtual image information generated by the first HUD. Since the real scene information and the virtual image information are diverse, and the correlation between the two is also designed in multiple ways, if the two are directly matched, the matching complexity is high and the calculation amount is large. For example, some virtual images need to be accurately overlaid on the real scene information (i.e., the position needs to be adapted), and some virtual image information needs to have the same directional direction as the real scene information (i.e., the direction needs to be adapted) but does not require accurate overlay on the real scene information.

[0011] In the above method, the testing device can extract features of the two types of information. The features are abstract representations of attributes of elements, and can digitalize the information contained in the real scene and the virtual image. The matching is more efficient and accurate. Therefore, the testing device determines the fitting degree evaluation result of the HUD based on the features of the two types of information, which can objectively evaluate the fitting degree of the projected image of the HUD with the real driving environment, and provides a more accurate and quantitative evaluation standard, which helps to improve the performance and user experience of the HUD technology.

[0012] Optionally, the projected image comprises virtual image information and fixed projection information, the fixed projection information being used to indicate elements that are only presented in the driver's field of view, do not change with the movement of the test vehicle or do not change with the change of the real scene information, and the virtual image information is specifically elements in the projected image that are mapped in response to the real scene information, in other words, the virtual image information is information calculated based on the real scene information, wherein the elements can be specifically icons / images of the first HUD projection.

[0013] For example, the virtual image information is an AR icon of the first HUD projection.

[0014] Optionally, the virtual image information comprises a plurality of AR icons of the first HUD projection.

[0015] In general, current AR icons can be classified into two categories in terms of application attributes: one is an attribute indication type that is fixed relative to the position of the real scene information (such as a preceding vehicle), and the other is a type that is not fixed relative to the position of the real scene information, such as a cruise indication type that indicates the action trajectory / predicted driving route / cruise related to the current vehicle.

[0016] It should be understood that the cruise indication type described above includes cruise indication information in the case of manual driving and / or cruise indication information in the case of intelligent driving.

[0017] Optionally, the extraction of the first feature is specifically determined based on the application attributes of the AR icon, and the first features corresponding to AR icons of different attributes are different.

[0018] In one possible implementation, based on the first feature and the second feature, the fitting degree evaluation result of the first HUD is determined, including the following operations: the test device determines a first fitting degree of the first feature and the second feature, and the test device determines the fitting degree evaluation result of the first HUD based on the first fitting degree.

[0019] In the above-mentioned implementation, the specific steps of determining the HUD fitting degree evaluation result based on the first feature and the second feature are further clarified. By determining the first fitting degree of the first feature and the second feature, and obtaining the evaluation result based on the fitting degree, the evaluation process is more specific and operable. This step of refinement helps to improve the accuracy and reliability of the evaluation, and provides strong support for the optimization of the HUD technology.

[0020] Optionally, the fitting degree evaluation result of the first HUD comprises a plurality of fitting degrees, in other words, the fitting degree evaluation result described above is determined based on a plurality of fitting degrees (including the first fitting degree).

[0021] In one possible implementation, the first fitting degree is related to the positions of the first feature and the second feature.

[0022] In the above-mentioned embodiments, the first degree of fit is related to the positions of the first feature and the second feature, in other words, the determination manner of the first degree of fit is related to the positions of the first feature and the second feature. Optionally, the determination manner of the first degree of fit comprises determining based on the relative positional relationship of the first feature and the second feature. Exemplarily, the relative positional relationship comprises the distance between the first feature and the second feature.

[0023] Exemplarily, for the case that the above-mentioned virtual image information comprises attribute indication information, the relative positional relationship between the first feature corresponding to the attribute indication information and the second feature corresponding to the real scene information should be fixed. Based on this, the evaluation of the degree of fit between the attribute indication information and the corresponding real scene information can be realized.

[0024] This further clarifies one of the key elements of the degree of fit evaluation, the positional relationship. The accuracy of the positional relationship is crucial for evaluating the degree of fit between the HUD projection image and the real environment, which helps to improve the accuracy and practicality of the evaluation.

[0025] In a possible implementation, the first degree of fit is further related to the type of the virtual image information, and the type of the virtual image information comprises a cruise indication type.

[0026] In the above-mentioned embodiments, the first degree of fit is related to the type of the virtual image information, in other words, the determination manner of the first degree of fit corresponding to different types of virtual image information is different. In this embodiment, the determination manner of the degree of fit corresponding to the virtual image information of the cruise indication type and the virtual image information of other types is different, which makes the evaluation more comprehensive and detailed, and helps to meet the needs in different application scenarios.

[0027] In a possible implementation, the virtual image information comprises one or more of a regular icon, an irregular icon, etc. The first feature of the regular icon is the geometric center point of the regular icon, and the regular icon comprises one or more of an icon with a regular geometric shape and an icon with a regular geometric outer frame. The first feature of the irregular icon is a key element in the irregular icon, and the key element comprises one or more of a key line or a key point.

[0028] It can be understood that for the regular icons in the virtual image information, generally used to indicate the corresponding real scene information, for example, the AR icon indicating the following target is the horizontal line below the following target, or the triangular warning icon indicating the obstacle on the road, such icons can be classified as attribute indicating icons, which can also be called attribute indicating information. Based on the above, it can be known that the attribute indicating information includes one or more of the pedestrian identification, the obstacle identification, the front vehicle identification, or the lane line identification. In order to indicate clearly, the relative position of such icons to the corresponding real scene is generally fixed, for example, the horizontal line indicating the following target will move with the following target, so that the specific shape / structure of such icons does not affect the determination of the fitting degree when extracting the features, and whether the fitting is directly judged according to whether such icons are offset from the expected position, so such icons can directly use the geometric center point as the first feature, which is simple to extract, clear to indicate, and can reduce resource consumption during evaluation.

[0029] And for the irregular icons, they can play other roles in addition to attribute indication, such as the cruise indicating information mentioned above. Exemplarily, the cruise indicating information is indicated by an arrow icon. The role of such icons is cruise / guidance, and a single point cannot represent its role, so the corresponding features can be indicated by key elements.

[0030] In the above-mentioned embodiments, the possible implementations of the first features of the regular icons and the irregular icons included in the virtual image information are described in detail. By defining specific first features for different types of icons, the feature extraction process is more standardized and automated, thereby improving the efficiency and accuracy of evaluation.

[0031] In a possible implementation, the irregular icons include one or more of curve identification, arrow identification, or road surface cover identification. Exemplarily, the curve identification includes lane line identification and the like.

[0032] As a possible implementation, the first feature of the curve identification is one or more first key points of the curve identification, and the first key point is the tangent line of the curve identification and the connection point of the curve identification.

[0033] As another possible implementation, the first feature of the arrow identification is a second key point or a first key line of the arrow identification, the second key point includes a starting point and an ending point of the arrow identification, and the first key line is a ray from the starting point to the ending point.

[0034] As another possible implementation, the first feature of the road surface cover identification is a plurality of third key points of the road surface cover identification, and the plurality of third key points include a center point close to the boundary on one side of the vehicle and a center point away from the boundary on the other side of the vehicle.

[0035] The primary feature extraction method for irregular icons (such as curve markers, arrow markers, and road surface cover markers) has been further refined. These feature extraction methods are more in line with the needs of real-world application scenarios, making the evaluation results more accurate and reliable.

[0036] Optionally, the aforementioned third key points may also include one or more of the center point or tangent point of the boundary on both sides of the road surface covering mark. It is understood that the center point is generally used when the boundary is a straight line, and the tangent point is generally used when the boundary is a curve.

[0037] In one possible implementation, the first feature of the cruise indication information is a fourth key point or a second key line of the cruise indication information. The fourth key point includes one or more of the starting point and the ending point of the cruise indication information, and the second key line is a ray from the starting point to the ending point.

[0038] The aforementioned cruise control information serves the purpose of cruise / guidance. A single point cannot represent its function; therefore, key elements indicating corresponding features can be used. Thus, the starting and ending points, or a ray from the starting to the ending point, representing the direction, are used as the first feature point. This is simple to extract, provides clear indication, and reduces resource consumption during evaluation. In the aforementioned implementation, defining the first feature of the cruise control information provides a concrete basis for evaluating the HUD's performance in cruise control. It can be understood that the function of cruise control information is to indicate the driving direction to the test vehicle.

[0039] This implementation method makes the evaluation process more comprehensive and detailed.

[0040] In one possible implementation, determining a first degree of fit between the first feature and the second feature, particularly when the first degree of fit is determined based on the positions of the first feature and the second feature, may include the following operation: The testing device determines the first degree of fit between the first feature and the second feature based on the first feature, the second feature, and a starting point from the driver's perspective. The first degree of fit measures the fit deviation between real-world information and virtual image information, and is a value based on the angle formed by the first feature and the second feature relative to the starting point from the driver's perspective. The testing device determines the first degree of fit between the first feature and the second feature based on the first degree of fit.

[0041] In the aforementioned implementation, a method for determining the first degree of fit based on a first feature, a second feature, and a starting point from the driver's perspective is proposed. This method measures the fit deviation between real-world information and virtual image information by calculating the included angle, thereby achieving a quantitative assessment of the degree of fit. The introduction of this step makes the evaluation process more objective and accurate, contributing to improving the overall performance of HUD technology.

[0042] It can be understood that the starting point of the driver's perspective is specifically the eye point of the driver, the first image includes the field of view of the driver's perspective, and therefore the calculation of the above degree of fit is to convert the relative positional relationship between the first feature and the second feature into an angular relationship in a three-dimensional space. Considering that different drivers have different postures and heights, the positions between the first feature and the second feature directly observed by the driver are different on a two-dimensional plane. Therefore, using the positional relationship (angular relationship) in a three-dimensional space can more accurately and objectively measure the fitting deviation between the real scene information and the virtual image information.

[0043] In a possible implementation, the virtual image information is a lane line mark, the first feature is one or more first key points of the lane line mark, and the first degree of fit between the first feature and the second feature is determined based on the first feature, the second feature, and the starting point of the driver's perspective, including the following operations: the testing device determines a second degree of fit between the first feature and the second feature based on the first feature, the second feature, and the starting point of the driver's perspective, and the second degree of fit includes multiple. The testing device determines a third degree of fit with the maximum value in the multiple second degrees of fit as the first degree of fit.

[0044] Generally, the lane line mark is generally used to prompt the driver about the lane line on the road, for example, prompting the driver that there is an approaching vehicle behind the lane line, and the driver should not drive beyond the lane line. Therefore, the lane line mark is generally coincident with the actual lane line of the road. It can be understood that the lane line mark is a special case of virtual image information. Specifically, the relative position between the lane line mark and the real scene information should be fixed. Therefore, the corresponding first feature should be the geometric center point of the mark. However, in actual applications, the lane line mark may partially coincide with the actual lane line or partially not coincide with the actual lane line. For example, the lane line mark itself is a curve. The curved lane line mark can be applied to a scenario in which the actual lane line of the road itself is a curve or the actual lane line of the road is a straight line but the projected lane line mark is a curve. Therefore, it is obviously unreasonable to directly identify the first feature as the geometric center point of the lane line mark. Therefore, there are multiple first key points of the lane line mark.

[0045] Optionally, the multiple first key points are the connection points of the tangent of the lane line mark and the lane line mark itself. Correspondingly, there are multiple first features, and therefore there are multiple second degrees of fit.

[0046] This embodiment is particularly aimed at the case in which there are multiple second degrees of fit. The third degree of fit with the maximum value in the multiple second degrees of fit is identified as the first degree of fit. The third degree of fit represents that the corresponding first key point / first feature and the second feature are farthest apart in the three-dimensional space, and the fitting deviation is the largest. Therefore, the third degree of fit as the first degree of fit of the lane line mark meets the accuracy requirement of the fitting degree evaluation.

[0047] In the above-mentioned embodiments, for the case of lane line identification of virtual image information, a specific method for determining the first fitting degree is proposed. By calculating the maximum value of the plurality of second fitting degrees as the first fitting degree, the evaluation result is more in line with the needs of actual application scenarios. This embodiment helps to improve the accuracy and reliability of HUD technology in lane line identification.

[0048] In a possible implementation, the greater the value of the first fitting degree, the lower the fitting degree.

[0049] In the above-mentioned embodiments, the relationship between the first fitting degree and the fitting degree is clarified. By stipulating that the greater the value of the first fitting degree, the lower the fitting degree, the evaluation result is more intuitive and easy to understand. This embodiment helps to improve the accuracy and user-friendliness of the evaluation.

[0050] In a possible implementation, the first fitting degree includes parameters such as the first horizontal fitting degree and the first vertical fitting degree.

[0051] In the above-mentioned embodiments, the first fitting degree is subdivided into the first horizontal fitting degree and the first vertical fitting degree, which further enriches the evaluation dimension of the fitting degree. By considering the fitting degree in both horizontal and vertical directions, the evaluation result is more comprehensive and accurate, which helps to improve the performance of HUD technology in different directions.

[0052] Optionally, the first horizontal fitting degree and the first vertical fitting degree are more applicable to the case where the specific shape of the virtual image information is a line.

[0053] In a possible implementation, the second feature of the real scene information is determined based on the target projection position of the virtual image information, and the target projection position is the position where the virtual image information is projected.

[0054] In the above-mentioned embodiments, a method for determining the second feature of the real scene information based on the target projection position of the virtual image information is proposed. This method makes the evaluation process more flexible and adaptable, and can cope with changes in demand in different scenarios. In addition, this embodiment helps to improve the performance of HUD technology in complex environments.

[0055] In a possible implementation, the first fitting degree is determined based on the target condition satisfied by the virtual image information.

[0056] In the above-mentioned embodiments, the first fitting degree is determined by defining the target condition satisfied by the virtual image information, which further refines the evaluation standard. By considering the specific needs of virtual image information in actual applications (such as the accuracy and continuity of cruise indication information), the evaluation result is more in line with the needs of actual application scenarios.

[0057] In a possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes an arrow mark or the cruise instruction information is an arrow mark, and the target condition includes at least one of the following: Condition 1, a cutoff point of the arrow mark is located in a target road, and the target road includes a target lane. Condition 2, the cutoff point of the arrow mark is located in the target lane. Condition 3, an extension line extending in a direction indicated by the arrow mark intersects the target road, and a distance between an intersection point and the vehicle is less than a first distance. Condition 4, the extension line extending in the direction indicated by the arrow mark intersects the target lane, and a distance between the intersection point and the vehicle is less than a second distance. Of course, the target condition can include more or fewer conditions in the specific implementation.

[0058] In the above-mentioned implementation, in the case where the cruise instruction information is an arrow mark, in order to ensure the cruise / guiding effect of the arrow mark, the corresponding target condition is set to a condition related to the cruise / guiding effect.

[0059] Optionally, the target condition includes a condition for indicating a direction of the vehicle from an actual feeling of the driver. It can be understood that the target condition includes conditions corresponding to multiple fitting levels, such as a cutoff point located in a target road and a cutoff point located in a target lane. It can be obviously seen that the guiding effect of the cutoff point located in the target lane is stronger than that of the cutoff point located in the target road.

[0060] Optionally, the target condition corresponding to the cruise instruction information should also include that the mark is not misleading.

[0061] In a possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes a road surface covering mark or the cruise instruction information is a road surface covering mark, and the target condition includes at least one of the following: Condition 1, a center point of a boundary close to a side of the vehicle is located in front of the vehicle. Condition 2, a center point of a boundary away from the side of the vehicle is located in a target road, the target road is a road to be traveled by the vehicle, and the target road includes a target lane. Condition 3, the center point of the boundary away from the side of the vehicle is located in the target lane. Condition 4, the road surface covering mark has a natural shape. Condition 5, the road surface covering mark is continuous and changes smoothly. Of course, the target condition can include more or fewer conditions in the specific implementation.

[0062] In a possible implementation, the first fitting degree is determined based on the target condition met by the virtual image information and a first scene, and the first scene belongs to a scene supported by the first HUD.

[0063] In the above-mentioned implementation, the first fitting degree is evaluated in combination with the scene supported by the HUD. This implementation makes the evaluation process more specific and targeted, and can customize the evaluation for different scenarios. The introduction of this step helps to improve the adaptability and accuracy of the HUD evaluation method in different scenarios.

[0064] In a possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes an arrow mark, the first scene includes a target road curvature greater than 0, and the target condition further includes consistency with the road curve trend of the target road, and the target road is a road to be traveled by the estimated vehicle.

[0065] In a specific scenario, such as the scenario where the target road curvature is greater than 0 or the scenario of a forked intersection, the cruise instruction should clearly indicate the corresponding road and cannot be misleading. Except that the current lane of the vehicle is a straight forward lane (the curvature is approximately equal to 0), but the target road is a forked intersection, which may have a curvature, i.e., the curvature of the target road is greater than 0, then the cruise instruction information only needs to indicate the position of the target road, or the trend of the road curve of the target road is consistent with the target road, even if the cut-off point does not fall within the target road / lane, it still plays a guiding role, therefore, in such scenarios, the target condition further includes consistency with the road curve trend of the target road.

[0066] Optionally, the above-mentioned "consistency with the road curve trend of the target road" can be replaced by "curvature equal to or close to the target road"

[0067] In a possible implementation, the first fitting degree is determined based on the number of target conditions satisfied by the virtual image information. The more target conditions satisfied by the virtual image information, the higher the fitting degree.

[0068] In the above-mentioned implementation, a method for determining the first fitting degree based on the number of target conditions satisfied by the virtual image information is proposed. By counting the number of target conditions satisfied by the virtual image information, the overall performance and performance of the HUD technology can be intuitively reflected. This implementation helps to improve the accuracy and comprehensiveness of the fitting evaluation, and provides strong support for the optimization of the HUD technology.

[0069] In a possible implementation, the method further includes the following steps: the testing device constructs a test scene database, the test scene database includes a plurality of test scenes, any two test scenes in the plurality of test scenes differ in at least one test factor in a plurality of test factors, and the plurality of test factors include one or more of a road static factor, a dynamic traffic participant factor, and an environmental factor.

[0070] The testing device acquires the first image, including the following steps: the testing device acquires the first image based on a first scene in a plurality of test scenes in a test scene database.

[0071] In the above-mentioned embodiments, a test scene database is constructed, and the first image is acquired based on the database for testing. It can be understood that the plurality of test scenes described above can be obtained by organically combining a plurality of test factors based on a traversal manner, and the first scene can be any scene in the test scene database.

[0072] This embodiment makes the testing process more systematic and standardized, and can cover more possible test scenes and factors. By constructing a comprehensive test scene database, the performance of the HUD technology in different scenes can be more effectively evaluated, providing strong support for the optimization of the technology.

[0073] In a possible implementation, the plurality of test factors further include a test vehicle self factor, and the test vehicle self factor includes one or more of vehicle basic information, position information, or motion state information, and the motion state information includes vehicle speed.

[0074] For example, in part or all of the test scenes, there are requirements for the speed of the test vehicle, for example, in the same road scene, the speed of the test vehicle should be (40±10) km / h and (80±10) km / h respectively.

[0075] In a possible implementation, the road static factor includes one or more of intersection shape information, road geometry information, road surface condition information, landmark information, or lane line condition information; the dynamic traffic participant factor includes one or more of relative position relationships between motor vehicles, non-motor vehicles, pedestrians and the test vehicle; and the environmental factor includes one or more of time, brightness, or weather.

[0076] In the above-mentioned embodiments, the test factors are further refined into road static factors, dynamic traffic participant factors, environmental factors, and test vehicle self factors, and the like. This embodiment makes the testing process more comprehensive and detailed, and can more accurately reflect the performance of the HUD technology in different scenes. In addition, this embodiment also helps to improve the accuracy and reliability of the test results.

[0077] In a possible implementation, the first image includes a second image and a third image, the second image and the third image are images spaced by a first time in the first image, the first fitting degree includes a second fitting degree and a third fitting degree, the second fitting degree is a fitting degree of virtual image information and real scene information in the second image, and the third fitting degree is a fitting degree of virtual image information and real scene information in the third image; based on the first fitting degree, the fitting degree evaluation result of the first HUD is determined, including the following operations: the test device determines the fitting degree evaluation result of the first HUD based on the second fitting degree and the third fitting degree, and the fitting degree evaluation result includes a jitter evaluation result.

[0078] In the above-mentioned implementation, a method for determining the fitting degree evaluation result of the first HUD based on the second image and the third image is proposed, wherein the second image and the third image are images spaced by a certain time. This implementation enables the evaluation process to consider the influence of time factors on the fitting degree of the HUD, thereby more comprehensively evaluating the performance and stability of the HUD technology. In addition, by introducing the jitter evaluation result (including the jitter amplitude and the jitter frequency), the performance capability of the HUD technology in a dynamic scene can be further quantified.

[0079] In a possible implementation, the jitter evaluation result includes one or more of the jitter amplitude or the jitter frequency, the jitter amplitude is used to indicate the maximum fluctuation degree of the fitting degree corresponding to the first image, and the jitter amplitude includes one or more of a single-cycle jitter amplitude or a jitter distribution dispersion degree; the single-cycle jitter amplitude includes a first jitter amplitude obtained based on the second fitting degree and the third fitting degree, and the second fitting degree and the third fitting degree are respectively a maximum value and a minimum value in the first time.

[0080] In a possible implementation, based on the first feature, the second feature, and the starting point of the driver's perspective, the first fitting degree between the first feature and the second feature is determined, including the following operations: the test device establishes a first coordinate system and a second coordinate system, the first coordinate system is related to the first image, and the origin of the first coordinate system is the top left corner vertex of the image; the second coordinate system is a three-dimensional coordinate system, and the origin of the second coordinate system is the starting point of the driver's field of view; the first image is obtained through an image acquisition device, and the center point of the lens position of the image acquisition device is the starting point of the driver's field of view. The test device determines a first coordinate of the first feature in the first coordinate system and a second coordinate of the second feature in the first coordinate system. The test device converts the first coordinate and the second coordinate into a third coordinate and a fourth coordinate respectively based on the intrinsic information and the extrinsic information of the image acquisition device, and the first coordinate system and the second coordinate system, wherein the third coordinate is the coordinate of the first feature in the second coordinate system, and the fourth coordinate is the coordinate of the second feature in the second coordinate system. The test device determines the first fitting degree between the first feature and the second feature based on the third coordinate and the fourth coordinate.

[0081] In the above-mentioned embodiments, the specific method of determining the first fit degree based on the first feature, the second feature, and the starting point of the driver's perspective is described in detail. By establishing the first coordinate system and the second coordinate system, and using the internal parameter information and the external parameter information of the image acquisition device for coordinate transformation, the accurate description of the first feature and the second feature in the three-dimensional space is realized. This embodiment makes the evaluation process more scientific and accurate, and provides strong support for HUD effect evaluation.

[0082] In a possible implementation, the first HUD is a windshield HUD or an augmented reality (AR) HUD.

[0083] In a possible implementation, the first image is obtained by the following operations: the testing device obtains the first image based on the first video, the first video includes a plurality of images, and the first image is an image including virtual image information and real scene information in the plurality of images.

[0084] In the above-mentioned embodiments, the specific type of the first HUD (such as a windshield HUD or an augmented reality (AR) HUD) is specified, and the specific method of obtaining the first image (such as extraction based on the first video) is pointed out. This embodiment makes the testing method more specific and operable, and can be directly applied to different types of HUD technology and testing scenarios. In addition, by providing specific testing steps and methods, it is helpful to promote the standardization and normalization development of HUD technology.

[0085] In a second aspect, the application provides a testing device, which can be used to implement the first aspect or any one of the methods in the first aspect, and includes corresponding functional modules for implementing the steps in the above methods. The functions can be realized by hardware, or realized by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions.

[0086] In a possible implementation, the testing device can include a communication unit and a processing unit. The communication unit is configured to obtain a first image, the first image being an image from the perspective of a driver of a test vehicle, the test vehicle being provided with a first HUD, the first HUD being configured to generate a projected image, the projected image being located in the driver's perspective of the test vehicle. The first image includes real scene information and virtual image information, the real scene information being an element in the real driving environment of the test vehicle, and the virtual image information being an element related to the real scene information in the projected image.

[0087] The processing unit is configured to obtain a first feature of the virtual image information and a second feature of the real scene information based on the first image, the first feature being used to represent an attribute of the virtual image information, and the second feature being used to represent an attribute of the real scene information, the attribute including a position and / or a direction; the virtual image information includes cruise instruction information and / or attribute instruction information, the attribute instruction information including one or more of a pedestrian identifier, an obstacle identifier, a front vehicle identifier, or a lane line identifier;

[0088] The processing unit is configured to obtain a first feature of the virtual image information and a second feature of the real scene information based on the first image, the first feature being used to represent an attribute of the virtual image information, and the second feature being used to represent an attribute of the real scene information, the attribute including a position and / or a direction; the virtual image information includes cruise instruction information and / or attribute instruction information, the attribute instruction information including one or more of a pedestrian identifier, an obstacle identifier, a front vehicle identifier, or a lane line identifier;

[0089] In a possible implementation, the processing unit is configured to determine the fitting degree evaluation result of the first HUD based on the first feature and the second feature, and specifically configured to: determine a first fitting degree of the first feature and the second feature, and determine the fitting degree evaluation result of the first HUD based on the first fitting degree.

[0090] In a possible implementation, the first fitting degree is related to the positions of the first feature and the second feature.

[0091] In a possible implementation, the first fitting degree is further related to a type of the virtual image information, and the type of the virtual image information includes a cruise instruction type.

[0092] In a possible implementation, the virtual image information includes a regular icon and / or an irregular icon.

[0093] The first feature of the regular icon is a geometric center point of the regular icon, and the regular icon includes an icon with a regular geometric shape and an icon with a regular geometric outer frame.

[0094] The first feature of the irregular icon is a key element in the irregular icon, and the key element includes a key line and / or a key point.

[0095] In a possible implementation, the irregular icon includes one or more of a curve identifier, an arrow identifier, or a road surface cover identifier, and the curve identifier includes a lane line identifier.

[0096] The first feature of the curve identifier is one or more first key points of the curve identifier, and the first key point is a connection point of a tangent line of the curve identifier and the curve identifier.

[0097] The first feature of the arrow identifier is a second key point or a first key line of the arrow identifier, the second key point includes a start point and an end point of the arrow identifier, and the first key line is a ray from the start point to the end point.

[0098] The first feature of the road cover mark is a plurality of third key points of the road cover mark, and the plurality of third key points include a center point close to a boundary on a side of the vehicle and a center point away from the boundary on the side of the vehicle.

[0099] In a possible implementation, the virtual image information includes cruise instruction information, and a first feature of the cruise instruction information is a fourth key point or a second key line of the cruise instruction information, the fourth key point includes a start point and an end point of the cruise instruction information, and the second key line is a ray from the start point to the end point.

[0100] In a possible implementation, the first fitting degree of the first feature and the second feature is determined based on the first fitting degree.

[0101] The first fitting degree of the first feature and the second feature is determined based on the first fitting degree.

[0102] The first fitting degree of the first feature and the second feature is determined based on the first fitting degree.

[0103] In a possible implementation, the virtual image information is a lane line mark, the first feature is one or more first key points of the lane line mark, and the first fitting degree of the first feature and the second feature is determined based on the first fitting degree.

[0104] The second fitting degree between the first feature and the second feature is determined based on the first feature, the second feature, and the start point of the driver's perspective.

[0105] The third fitting degree with the maximum value in the plurality of second fitting degrees is determined as the first fitting degree.

[0106] In a possible implementation, the greater the value of the first fitting degree, the lower the fitting degree.

[0107] In a possible implementation, the first fitting degree includes a first horizontal fitting degree and a first vertical fitting degree.

[0108] In a possible implementation, the second feature of the real scene information is determined based on a target projection position of the virtual image information, and the target projection position is a position where the virtual image information is projected.

[0109] In a possible implementation, the first fitting degree is determined based on a target condition satisfied by the virtual image information.

[0110] In a possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes an arrow mark, and the target condition includes at least one of the following:

[0111] The end point of the arrow mark is located in a target road, and the target road includes a target lane;

[0112] The end point of the arrow mark is located in the target lane;

[0113] An extension line extending in the direction indicated by the arrow mark intersects the target road, and the distance between the intersection point and the vehicle is less than a first distance;

[0114] The extension line extending in the direction indicated by the arrow mark intersects the target lane, and the distance between the intersection point and the vehicle is less than a second distance.

[0115] In a possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes a road surface cover mark, and the target condition includes at least one of the following:

[0116] The center point of the boundary close to the vehicle side is located in front of the vehicle;

[0117] The center point of the boundary away from the vehicle side is located in a target road, and the target road is a road on which the vehicle is estimated to travel, and the target road includes a target lane;

[0118] The center point of the boundary away from the vehicle side is located in the target lane;

[0119] The road surface cover mark is natural in shape;

[0120] The road surface cover mark is continuous and smooth.

[0121] In a possible implementation, the first fitting degree is determined based on the target condition met by the virtual image information and a first scene, and the first scene belongs to a scene supported by the first HUD.

[0122] In a possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes an arrow mark, the first scene includes that the curvature of a target road is greater than 0, and the target condition further includes consistency with the road curve trend of the target road, and the target road is a road on which the vehicle is estimated to travel.

[0123] In a possible implementation, the first fitting degree is determined based on the number of target conditions met by the virtual image information, and the more target conditions met by the virtual image information, the higher the fitting degree.

[0124] In a possible implementation, the processing unit is further configured to:

[0125] The test scene database is constructed, and the test scene database includes a plurality of test scenes. Any two test scenes in the plurality of test scenes are different in at least one test factor. The plurality of test factors include a road static factor, a dynamic traffic participant factor, and an environment factor.

[0126] Further, the first image is acquired. The communication unit is specifically configured to:

[0127] The first image is acquired based on a first scene in the plurality of test scenes in the test scene database.

[0128] In a possible implementation, the plurality of test factors further include a test vehicle self factor. The test vehicle self factor includes one or more of vehicle basic information, position information, or motion state information. The motion state information includes a vehicle speed.

[0129] In a possible implementation, the road static factor includes one or more of intersection shape information, road geometry information, road surface condition information, landmark information, or lane line condition information.

[0130] The dynamic traffic participant factor includes one or more of relative position relationships between a motor vehicle, a non-motor vehicle, a pedestrian, and the test vehicle.

[0131] The environment factor includes one or more of a time, brightness, or weather.

[0132] In a possible implementation, the first image includes a second image and a third image. The second image and the third image are images that are separated from each other by a first time in the first image. The first fitting degree includes a second fitting degree and a third fitting degree. The second fitting degree is a fitting degree of virtual image information and real scene information in the second image. The third fitting degree is a fitting degree of virtual image information and real scene information in the third image.

[0133] The fitting degree evaluation result of the first HUD is determined based on the first fitting degree. The processing unit is specifically configured to:

[0134] The fitting degree evaluation result of the first HUD is determined based on the second fitting degree and the third fitting degree. The fitting degree evaluation result includes a jitter evaluation result.

[0135] In a possible implementation, the jitter evaluation result includes a jitter amplitude and a jitter frequency. The jitter amplitude is used to indicate a maximum fluctuation degree of the fitting degree corresponding to the first image. The jitter amplitude includes a single-period jitter amplitude and a jitter distribution dispersion degree.

[0136] The single-cycle jitter amplitude includes a first jitter amplitude obtained based on a second fitting degree and a third fitting degree, the second fitting degree and the third fitting degree being maximum and minimum values in the first time respectively.

[0137] In a possible implementation, based on the first feature, the second feature, and the starting point of the driver's perspective, the processing unit is specifically configured to determine the first fitting degree between the first feature and the second feature in terms of:

[0138] A first coordinate system and a second coordinate system are established, the first coordinate system is related to the first image, and the origin of the first coordinate system is the top left corner vertex of the image; the second coordinate system is a three-dimensional coordinate system, and the origin of the second coordinate system is the starting point of the driver's field of view; the first image is acquired by an image acquisition device, and the center point of the lens position of the image acquisition device is the starting point of the driver's field of view;

[0139] The first feature is determined in a first coordinate of the first coordinate system, and the second feature is determined in a second coordinate of the first coordinate system;

[0140] Based on the internal parameter information and the external parameter information of the image acquisition device, and the first coordinate system and the second coordinate system, the first coordinate and the second coordinate are converted into a third coordinate and a fourth coordinate respectively, the third coordinate being the coordinate of the first feature in the second coordinate system, and the fourth coordinate being the coordinate of the second feature in the second coordinate system;

[0141] Based on the third coordinate and the fourth coordinate, the first fitting degree between the first feature and the second feature is determined.

[0142] In a possible implementation, the first HUD is a windshield HUD or an augmented reality (AR) HUD.

[0143] In a possible implementation, the first image is acquired, and the communication unit is specifically configured to:

[0144] Based on the first video, the first image is acquired, the first video including a plurality of images, and the first image being an image including virtual image information and real scene information in the plurality of images.

[0145] In a third aspect, the application provides a computing device for implementing the first aspect or any one of the methods in the first aspect, including corresponding functional modules for implementing the steps in the above methods. The functions can be implemented by hardware or by hardware executing corresponding software. The hardware or software includes one or more modules corresponding to the above functions. Optionally, the computing device is the test device of the second aspect or includes the test device of the second aspect.

[0146] In a possible implementation, the computing device can include a transceiver and a processor. The processor can be configured to support the computing device to perform the corresponding functions of the computing device shown above, and the transceiver is configured to support the communication between the computing device and other devices. The transceiver can be a separate receiver, a separate transmitter, a transceiver with integrated transceiver function, or an interface circuit. Optionally, the computing device can further include a memory, which can be coupled with the processor and stores the necessary program instructions and data of the computing device.

[0147] In a fourth aspect, the present application provides a test system, which includes a computing device, an image acquisition device and a test vehicle. The computing device can be the computing device of the third aspect, or a device including the test device of the second aspect, and can be used to execute the method of the first aspect or any possible implementation of the first aspect.

[0148] The image acquisition device is installed on the corresponding line of the lower viewing angle (LDA) of the driver of the main driving position of the test vehicle, and the lens of the image acquisition device faces outward of the test vehicle. The image acquisition device is used to acquire the video or image of the driver's viewing angle of the test vehicle.

[0149] The test vehicle includes a first HUD, and the first HUD is used to generate a projection image. The position of the projection image is located in the driver's viewing angle of the test vehicle.

[0150] In a possible implementation, the computing device is connected with the test vehicle, and the computing device is used to acquire the related data of the first HUD of the test vehicle in the test condition.

[0151] In a fifth aspect, the present application provides a chip, which includes a processor and a communication interface. The communication interface is used to receive and / or send data, and / or the communication interface is used to provide input and / or output for the processor. The processor is used to implement the method of the first aspect or any possible implementation of the first aspect.

[0152] In a sixth aspect, the present application provides a computer readable storage medium, which stores a computer program or instructions. When the computer program or instructions are executed by a device, the computing device executes the method of the first aspect or any possible implementation of the first aspect.

[0153] In a seventh aspect, the present application provides a computer program product, which includes a computer program or instructions. When the computer program or instructions are executed by a computing device, the method of the first aspect or any possible implementation of the first aspect is implemented.

[0154] The beneficial effects of the second to seventh aspects mentioned above can be referred to the beneficial effects corresponding to the methods in the first aspect or any possible implementation of the first aspect. Attached Figure Description

[0155] The accompanying drawings used in the description of the embodiments will be briefly introduced below.

[0156] Figure 1 This is a schematic diagram of a HUD imaging effect;

[0157] Figure 2 A schematic diagram of the architecture of a testing system provided in an embodiment of this application;

[0158] Figure 3 A schematic diagram of a testing system provided in an embodiment of this application;

[0159] Figure 4 A flowchart illustrating a testing method provided in an embodiment of this application;

[0160] Figure 5 A schematic diagram of a first image provided for an embodiment of this application;

[0161] Figure 6 A schematic diagram of a projected image provided in an embodiment of this application;

[0162] Figure 7 A schematic diagram of virtual image information of an attribute indicator class provided in an embodiment of this application;

[0163] Figure 8 A schematic diagram of cruise instruction information provided in an embodiment of this application;

[0164] Figure 9 A schematic diagram illustrating yet another type of cruise instruction information provided in an embodiment of this application;

[0165] Figure 10 A schematic diagram illustrating the relative positional relationship between a dynamic traffic participant and a test vehicle, provided for an embodiment of this application;

[0166] Figure 11 This is a schematic diagram illustrating an application scenario of a testing method provided in an embodiment of this application;

[0167] Figure 12 A schematic diagram of a rule icon and a corresponding first feature provided for an embodiment of this application;

[0168] Figure 13 A schematic diagram illustrating a common identifier and its corresponding first feature, provided for embodiments of this application;

[0169] Figure 14 A schematic diagram illustrating a fit according to an embodiment of this application;

[0170] Figure 15 A second fitting degree provided for an embodiment of the present application;

[0171] Figure 16 A first coordinate system provided for an embodiment of the present application;

[0172] Figure 17 A second coordinate system provided for an embodiment of the present application;

[0173] Figure 18 An arrow identification in different scenarios provided for an embodiment of the present application;

[0174] Figure 19 A road surface cover identification provided for an embodiment of the present application;

[0175] Figure 20 A shaking amplitude provided for an embodiment of the present application;

[0176] Figure 21 A structure schematic diagram of a test device provided for an embodiment of the present application;

[0177] Figure 22 A structure schematic diagram of a computing device provided for an embodiment of the present application. DETAILED DESCRIPTION

[0178] The embodiments of the present application will be further described in detail below with reference to the accompanying drawings.

[0179] In order to facilitate understanding, the following examples are given to illustrate part of the concepts related to the embodiments of the present application for reference. As follows:

[0180] First, some terms in the present application are explained to facilitate understanding by those skilled in the art.

[0181] (1) Augmented reality (AR), also known as mixed reality. It applies virtual information to the real world through computer technology, and real-time superimposes virtual objects and real environment in the same picture or space. Augmented reality provides information different from what humans can perceive in general. It not only shows the information of the real world, but also displays virtual information at the same time, and the two kinds of information complement and superimpose each other.

[0182] (2) Head up display (HUD), also known as head-up display, is a display device that projects images into the driver's field of view in front. HUD was first applied to military aircraft, aiming to reduce the frequency of the pilot's need to look down to check the instrument. At the beginning, HUD can project driving-related information on the pilot's helmet through optical principles, so that the pilot can focus on the flight indicators and receive ground-transmitted information while ensuring normal driving, thereby improving the safety and convenience of driving. Today, HUD has begun to be applied to cars. At the initial design stage, the driving force for applying HUD is to ensure driving safety, so that the driver does not need to lower his head to shift his gaze to the instrument panel or the center control during driving, avoiding the switching of the eye focal length between the instrument panel or the center control and the road surface, and reducing the time of crisis response. Therefore, the information projected by the HUD at the initial design stage is mainly the driving status indicators of the car, such as the speed, fuel level and other simple information displayed on the instrument panel.

[0183] The common HUD on the market can include components such as a projector, a mirror (or called secondary mirror) and a projection mirror (or called primary mirror). The imaging principle of HUD is similar to slide projection, which projects images onto the windshield of the car so that the driver can obtain the image in front of the field of view. Specifically, light information can be first emitted by the projector, reflected by the mirror to the projection mirror, and then reflected by the projection mirror to the windshield. The human eye sees a virtual image about 2-2.5 meters in front of the eye, giving the feeling that the information is floating in front of the road.

[0184] As mentioned above, HUD uses optical reflection principle to project and display information such as overspeed warning, vehicle condition detection, fuel consumption, speed, etc. on the windshield, which can make the driver's attention focused on the road in front, and achieve active driving safety. At the same time, it can also reduce the delay and discomfort caused by the constant adjustment of eye focal length.

[0185] Further, designers hope to achieve the goal of intelligent driving through HUD and give it more functions. Based on this, in recent years, augmented reality head up display (AR-HUD) has been proposed. AR-HUD has stronger intuitiveness for users, which can fuse the image projected by HUD with real road information, enhance the driver's acquisition of road information, for example, it can project and display some virtual arrows in real time to intuitively guide us forward, thereby avoiding the situation of driving through the intersection and distracting the driver's attention during driving.

[0186] Generally, HUD refers not only to a head-up display, but also to a head-up display system, which is a vehicle-mounted electronic image system for projecting virtual image information in the driver's forward field of view, mainly including an imaging medium and a head-up display. Exemplarily, the head-up display system can be a windshield head-up display system or an augmented reality head-up display system. The imaging medium of the augmented reality head-up display system is the front windshield glass, and it does not need to have an imaging medium. It combines augmented reality technology to establish a mapping relationship between the actual environment and the virtual image, so that the driver sees the head-up display system after combining the actual environment with the image. The augmented reality head-up display system is a type of windshield head-up display system.

[0187] (3) Virtual image: the image formed by the image projected by the optical-mechanical display of the head-up display, which can be directly seen by the driver. This image cannot be received by the light screen.

[0188] (4) Eyebox: the eyebox generally refers to the range in which the driver's eyes can see the entire display image. The general eyebox size is 130mmx50mm. The eyebox can refer to the union of the effective aperture solid angle of the virtual image corresponding to the driver's eyes, which specifies the maximum movement range of the human eye that can observe complete display information, and generally specifies relevant values in the horizontal and vertical directions. Exemplarily, due to the height of different drivers, the eyebox needs to meet the movement range of about ±50mm in the vertical direction. In the embodiments of the present application, the area in which the human eye can see a clear HUD virtual image within the eyebox range. If the human eye is aligned with the center of the eyebox, a complete and clear HUD virtual image can be obtained. When the eyes move left and right or up and down, at a certain point in each direction, the image will deteriorate until it is unacceptable, i.e., beyond the eyebox range. In the area beyond the eyebox, there may be problems such as image distortion, color error, and even no display.

[0189] Please refer to Figure 1 , Figure 1 is a schematic diagram of the imaging effect of a HUD. As Figure 1 indicated, the projection surface of the HUD projected onto the windshield of the vehicle can be located directly in front of the driver's field of view. As Figure 1 indicated, compared with the previous HUD, the HUD can not only project and display basic driving speed and vehicle power information, but also display AR cruise guidance arrows and other information, thereby assisting the driver to achieve more intelligent, comfortable and safe driving.

[0190] As described above, the HUD can realize AR cruising and AR early warning and the like through the projection image of the HUD. Optionally, functions such as distance early warning, line early warning, traffic light monitoring, early lane change indication, pedestrian early warning, road sign display, lane deviation indication, front obstacle early warning, and driver state monitoring can be realized, and details are not described herein.

[0191] Currently, the test scheme for the AR effect of the HUD in the industry mostly relies on subjective judgment of the video observed by the driving vehicle. This method not only consumes time and effort, but also is easily affected by the personal experience and subjective feeling of the tester, resulting in that the accuracy and reliability of the evaluation result are questioned.

[0192] In order to solve the problem that there is no unified standard and no quantitative or qualitative evaluation scheme for the current HUD fitting detection, the embodiments of the present application provide a series of schemes, based on existing computing devices (such as mobile phones, vehicle machines, servers, etc.), various scenes are constructed through real scenes or software, so as to detect the fitting degree of the projection image of the HUD and the real scene in the various scenes, which can include obtaining image information including the HUD projection image corresponding to the various scenes, and evaluating the fitting degree of the related elements of the projection image in the image information and the related elements of the real scene road, to give an accurate and effective HUD image and real scene fitting degree evaluation scheme.

[0193] Further, considering that the real scene information and the virtual image information are various, and the association between them also has a variety of possible designs, if they are directly matched, the matching complexity is high and the calculation amount is large. For example, some virtual images need to be accurately overlaid on the real scene information (i.e. the position needs to be adapted), and some virtual image information needs to have the same guiding direction as the real scene information (i.e. the pointing needs to be adapted) but does not require to be accurately overlaid on the real scene information. In the above method, the test device can extract the features of the two kinds of information, the feature is an abstract representation of the attribute of the element, and the information contained in the real scene and the virtual image can be digitized, which is more efficient and accurate in matching and evaluation.

[0194] Based on this, the embodiments of the present application provide a test system, please see Figure 2 , Figure 2 The architecture schematic diagram of the test system provided by the embodiments of the present application, which can be applied to the HUD effect test on the real scene road, and can also be applied to the HUD effect test of various scenes built by software.

[0195] The test system comprises a computing device, an image acquisition device and a test vehicle.

[0196] The first HUD is arranged in the test vehicle. The first HUD is used to generate a projection image, and the position of the projection image is located in the driver's view angle of the test vehicle.

[0197] Optionally, the embodiments of the present application are described with the first HUD as an AR-HUD. It should be noted that the HUD virtual image generated by the AR-HUD can be projected in the driver's forward field of view. The main principle of the AR-HUD is to use multiple curved mirrors or flat mirrors to magnify and reflect the HUD virtual image generated by the picture generate unit (PGU) to a certain position outside the vehicle, that is, to the driver's forward field of view (eyebox range), so as to present the image of a certain distance (for example, 2 to 20 m) away from the road to the driver. The actual position of the HUD virtual image is determined by the optical system of the HUD.

[0198] According to the cruise demand in vehicle driving, the imaging distance of the AR-HUD reaches more than 7.5 meters, so that the HUD virtual image can be superimposed with the real scene of the object or the road surface to form an augmented reality effect, so that the driver can observe the real environment while obtaining prompt information, and there is no longer a visual blind area. It should be understood that if the AR-HUD only displays some vehicle speed and prompt information, it does not need to care too much about the fitting problem of the HUD virtual image, and if it involves cruise, advanced driving assistant system (ADAS) information, etc., it needs to obtain the accurate position of the HUD virtual image.

[0199] In theory, the higher the fitting degree of the cruise lane line and the related warning information projected by the AR-HUD to the actual road, the better, and there is no error at best. However, due to many technologies and actual road conditions, the actual fitting situation is uneven in the same HUD projection image in different scenes.

[0200] Optionally, the above-mentioned image acquisition device is arranged in the eyebox area, wherein the eyebox range is usually about 10 cm. The image acquisition device can be a camera, a camera, or other image / video acquisition sensors or sensor assemblies.

[0201] The above-mentioned test system can include one or more image acquisition devices. If the above-mentioned test system includes only one image acquisition device, the image acquisition device is arranged on the line corresponding to the downward viewing angle of the driver at the main driving position of the test vehicle. Illustratively, the image acquisition device is a camera, which includes at least one lens, and the center point of the at least one lens of the installed camera coincides with the center point of the driver's eyes, that is, the center point of the lens can be called the eye point. It can be understood that the image (such as the first image) acquired by the image acquisition device in the three-dimensional space has the eye point as the origin of the image acquired by the device.

[0202] In addition, if the test system includes multiple image acquisition devices, the multiple image acquisition devices can be arranged at different positions and all arranged in the eyebox region, for example, two image acquisition devices are arranged for image acquisition of the driver's two eyes.

[0203] Based on the above, the image acquisition device is used to acquire images in the field of view of the driver, which should include the projected image projected by the first HUD, so the lens of the image acquisition device is directed outward of the test vehicle. For reference, see Figure 3 , Figure 3 A schematic diagram of a test system according to an embodiment of the present application is provided.

[0204] Optionally, in the actual road test scenario, the image acquisition device can be fixed to the driver's head or the position where the driver's head is located by the fixing assembly.

[0205] In a possible implementation, the computing device is connected to the test vehicle, and the computing device is used to acquire the relevant data of the first HUD of the test vehicle in the test situation. Optionally, the relevant data of the first HUD is used for data backtracking in the simulation scenario built by software.

[0206] In a possible implementation, the image acquisition device is installed and calibrated before the test starts.

[0207] In a possible implementation, the image acquisition device is connected to the computing device, and the computing device acquires the video / image including the driver's field of view through the image acquisition device, and then analyzes the fitting degree of the real scene information and the virtual image information to obtain the fitting degree evaluation result corresponding to the first HUD of the test vehicle.

[0208] In a possible implementation, the test vehicle starts all the augmented display functions before the test, and the augmented display function can be an augmented reality related function or a HUD display effect enhancement function.

[0209] In a possible implementation, during the image acquisition process, the image acquisition device continuously acquires and records the virtual image in a virtual image acquisition reference posture.

[0210] In a possible implementation, the image acquisition device or the computing device selects the acquisition image with augmented display information.

[0211] In an optional implementation, the computing device includes an optical test module, a feature recognition module, a coordinate conversion module, and a fitting calculation module.

[0212] The optical testing module is configured to detect, calibrate and store optical parameters of the image capturing device and / or the first HUD before the test starts. Exemplarily, the optical parameters include one or more of a virtual image distance (VID), a field of view (FOV), a look over angle (LOA), and a look down angle (LDA). Optionally, the field of view includes a horizontal field of view (HFOV) and a vertical field of view (VFOV).

[0213] The feature recognition module is configured to determine corresponding features from the real scene information and the virtual image information in the first image. For example, the virtual image information corresponds to a first feature and the real scene information corresponds to a second feature. The detailed extraction process is described in the subsequent content and will not be repeated here. However, it should be understood that the first / second feature is extracted based on the attributes of the corresponding information, which includes position and / or direction. Optionally, the first features corresponding to the virtual image information of different attributes are different.

[0214] The coordinate conversion module is configured to convert the coordinates of the first feature and the second feature on the image (e.g., the first image captured by the image capturing device) to coordinates in a three-dimensional space. Exemplarily, the three-dimensional space corresponds to an eye point as the origin of the coordinate system.

[0215] The fitting calculation module is configured to calculate the fitting degree corresponding to the coordinates obtained by the coordinate conversion module, or to qualitatively evaluate the first feature and the second feature to obtain a fitting degree evaluation result corresponding to the first HUD.

[0216] Optionally, the computing device further includes a communication unit configured to receive the first image from the image capturing device. Optionally, the first image includes real scene information and a projection image projected by the first HUD. The projection image should include virtual image information, which is an element related to the real scene information in the projection image. For ease of understanding, the virtual image information is an AR icon, such as an AR cruise indication arrow, an AR cruise indication light carpet, and other cruise indication information.

[0217] Optionally, the computing device further includes a storage module configured to store images captured by the image capturing device and / or related data of the first HUD.

[0218] Optionally, the computing device and the image capturing device are the same device, i.e., the image capturing device is a built-in image capturing device of the computing device, such as a mobile phone, a tablet computer, etc. with a built-in camera.

[0219] In summary, the computing device can be a smartphone, a smart wearable device, a tablet computer, a notebook computer, a desktop computer, and an in-vehicle infotainment system, etc. with the above functions, and the embodiments of the present application do not make specific limitations. The computing device can also be a computer and a server, etc. with the above functions, such as a server, a server cluster composed of multiple servers, or a cloud computing service center, for example, a vehicle Internet service platform, etc. The embodiments of the present application do not make specific limitations.

[0220] The test vehicle can be a car, a truck, a motorcycle, a bus, a ship, a drone, an airplane, a helicopter, a lawn mower, an entertainment vehicle, an amusement park vehicle, a construction equipment, a trolleybus, a golf cart, a train, and a cart, etc. with a HUD, and the embodiments of the present application do not make specific limitations.

[0221] Based on the above content, please refer to Figure 4 , Figure 4 The flowchart of a test method provided by the embodiments of the present application. Optionally, the execution subject of the method can be a computing device in the test system involved in the above Figure 2 or Figure 3 , of course, for the convenience of understanding, the following still takes the test device as an example for description, which can be a computing device or a module in the computing device. The module can be a software module and / or a hardware module, such as a chip, a computer executable program, etc.

[0222] Figure 4 The test method shown includes one or more steps in steps S401 to S403. It should be understood that Figure 4 for the convenience of understanding the scheme, the order of steps S401 to S403 is used for description, but the embodiments of the present application do not make limitations on the execution order, execution time, execution times, etc. of the above one or more steps. Steps S401 to S403 are as follows:

[0223] Step S401: The test device acquires a first image.

[0224] The first image is an image under the driver's perspective of the test vehicle, and the test vehicle includes a first HUD, which is used to generate a projection image, and the position of the projection image is located in the driver's perspective of the test vehicle.

[0225] It should be understood that the first image is used to indicate the projection image (or the virtual image information in the projection image) of the first HUD and the real scene information of the actual road under the driver's perspective. In other words, the first image includes real scene information and virtual image information, and the real scene information is an element in the real driving environment of the test vehicle.

[0226] For example, the real scene information can be a pedestrian, other vehicles, road signs (such as traffic lights, signs, etc.), obstacles, and the like on the actual road. It can be understood that the real scene information can be any thing that can appear in the process of the vehicle (such as the test vehicle) driving on the road. For details, see Figure 5 , Figure 5 A schematic diagram of a first image is provided for an embodiment of the present application. In Figure 5 , the real scene information includes a bridge pier and a pedestrian driving a non-motor vehicle, the virtual image 1 corresponds to the real scene bridge pier, and is used to indicate the height of the bridge pier or the passable height under the bridge; the virtual image 2 corresponds to the pedestrian driving the non-motor vehicle. It should be noted that the virtual image is a concrete manifestation of the virtual image information, and is used to indicate the virtual image information of a certain appearance.

[0227] Alternatively, the real scene information can also refer to an element that interacts with the vehicle (such as the test vehicle) in the process of driving on the road. For example, the test vehicle can follow other vehicles (i.e., the following target) when performing lane cruising or intelligent driving functions, and the real scene information can also refer to the following target.

[0228] Further, the virtual image information is an element in the projection image that is related to the real scene information. The virtual image information is exemplarily described below taking the following target as an example. For the following target, part of the HUD can project an icon (such as a horizontal line below the following target, which is used to prompt the user that the vehicle is the current following target) indicating the following target. Then, the icon (such as the horizontal line below the following target) is an element related to the real scene information (the following target), i.e., the icon is the virtual image information. The virtual image information is specifically an element in the projection image that is mapped and echoed with the real scene information. In other words, the virtual image information is information calculated according to the real scene information, wherein the element can specifically refer to an icon / image projected by the first HUD.

[0229] Exemplarily, the virtual image information is an AR image projected by the first HUD, which can also be referred to as an icon.

[0230] It can be understood that the projection image projected by the first HUD not only includes the virtual image information, but also includes fixed projection information. The fixed projection information is used to indicate elements that are suspended in the driver's field of view and only present corresponding information, do not change with the movement of the test vehicle, or do not change with the change of the real scene information, such as the current speed information projected by the first HUD. For details, see Figure 6 , Figure 6 A schematic diagram of a projection image is provided for an embodiment of the present application.

[0231] For ease of understanding, the projection image can be understood as an image projected by the first HUD, such as Figure 6As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as Figure 6 As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as Figure 6 As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as

[0232] As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as Figure 6 As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as Figure 6 As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as

[0233] Generally, the current AR icon can be classified into two categories in terms of application attributes. One category is attribute indication type fixed with the relative position of the real scene information (such as the following target), and the other category is not fixed with the relative position of the real scene information, such as the cruise indication type indicating the action track / predicted driving route / cruise of the current vehicle.

[0234] It should be understood that the cruise indication type includes cruise indication information in the case of manual driving and / or cruise indication information in the case of intelligent driving.

[0235] As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as Figure 6 As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as

[0236] Figure 7 As shown in the dashed box in FIG. 1, the virtual image information can be understood as AR image information in the projected image, such as

[0237] The following will exemplarily describe the AR icon (virtual image information) of the two application attributes.

[0238] The first category is attribute indication type.

[0239] The attribute-indicating AR icon of the attribute-indicating type is specifically used to indicate an attribute in the out-of-vehicle scene, such as weather, season, geographical location, road condition, terrain, traffic signal, and the like, or a pedestrian, an obstacle, another vehicle, and the like.

[0240] Optionally, each out-of-vehicle scene and each AR icon can have one or more corresponding attributes, and if the virtual image information in the first image is of the attribute-indicating type, the attribute of the virtual image information can be the same (or matched) with the attribute of the first out-of-vehicle scene. The first out-of-vehicle scene includes real scene information. For example, Figure 7 Figure 7 An attribute-indicating virtual image information provided by an embodiment of the present application is shown in the following diagram. Figure 7 In the diagram, the scene that the vehicle is expected to pass through is a culvert (or a tunnel), and the attribute of the scene includes the height of the culvert (or the height of the tunnel). The virtual image information includes an AR icon indicating the height of the culvert, such as 3.5 m shown in the diagram. Figure 7 The real scene information related to the virtual image information is the entrance (or the edge of the entrance, a stone column, or an inner wall of the entrance, and the like) of the culvert (or the tunnel), and the relative position of the virtual image information and the real scene information should be fixed.

[0241] For another example, the attribute of the out-of-vehicle scene also includes the type of the obstacle, and in this case, the virtual image information also includes information indicating the type of the obstacle. For example, when the type of the obstacle is a non-motor vehicle, the virtual image information includes an AR icon indicating that the obstacle is a non-motor vehicle.

[0242] In some possible implementations, the attribute-indicating virtual image information can also include one or more of a pedestrian identifier, an obstacle identifier, a front vehicle identifier, or a lane line identifier. It should be understood that different identifiers are used to indicate different attributes of real scene information, such as a pedestrian identifier used to indicate the position of a pedestrian, an obstacle identifier used to indicate an obstacle and the position of the obstacle, a front vehicle identifier that can be the horizontal line indication of the above-mentioned following target, used to indicate the current following target, and a lane line identifier used to indicate the lane line on one side or both sides of the vehicle, used to indicate that there is an oncoming vehicle behind the vehicle and the vehicle is prohibited from crossing the lane line.

[0243] For example, a related identifier for a bicycle, referred to as a non-motor vehicle identifier herein, is used to indicate a non-motor vehicle on the road, and the relative position of this type of identifier and the non-motor vehicle is fixed. Figure 8

[0244] ​​It should be understood that the specific use of the above-mentioned various identifiers is only an example, and in actual application, the same identifier can play different indicating roles. For example, the lane line identifier can also be used to indicate the position of the lane line when the actual road lane line is unclear, thereby playing a prompting role. Considering the continuous development of AR technology in the future, the types and roles of AR icons will become more and more rich, and therefore the specific forms and roles of the attribute indicating type of virtual image information are not limited in the present application.

[0245] The second type is the cruise indicating type.

[0246] The virtual image information of the cruise indicating type can also be cruise indicating information. In order to facilitate understanding, the cruise indicating information is used instead of the virtual image information of the cruise indicating type in the following. The role of the cruise indicating information in the vehicle driving process is very crucial.

[0247] The cruise indicating information is used to indicate the direction or road of the pre-driving of the test vehicle. For example, refer to Figure 8 , Figure 8 A schematic diagram of the cruise indicating information provided by an embodiment of the present application.

[0248] In Figure 9 , the cruise indicating information is presented in the form of an arrow identifier, which is used to indicate that the current vehicle (test vehicle) should drive forward in the road indicated by the cruise map.

[0249] It should be understood that the cruise indicating information can also be presented in the form of a road surface covering identifier (such as a light carpet) in addition to the form of an arrow identifier. For details, refer to Figure 10 . It should be understood that the road surface covering identifier mainly plays a role of indicating the road. For example, in the vehicle lane changing scene, the road surface covering identifier covers the road on which the current vehicle (test vehicle) should drive, or even covers the lane on which the current vehicle should drive, thereby playing a role of indicating the road / lane.

[0250] Generally, multiple cruise indicating information will not appear in the image projected by the HUD. Basically, it is mainly to prevent the simultaneous appearance of multiple cruise indicating information from affecting the presentation of the cruise / guiding role, and possibly affecting the observation of the actual road by the driver.

[0251] The virtual image information of the two types can be applied to the method provided by the embodiments of the present application. In other words, the method provided by the embodiments of the present application can basically cover the AR icons that can be projected by the current HUD. It should be understood that in some scenarios, the AR icon can be displayed only in the case of intelligent driving of the vehicle, and in another part of the scenario, the AR icon can also be displayed in the case of manual driving of the vehicle (turning on the navigation). The related content of the cruise indication type and the cruise indication information involved in the embodiments of the present application covers the indication information related to cruise / navigation / guidance that can appear in the two driving scenarios of intelligent driving and manual driving.

[0252] Further, the virtual image information included in the first image can be the attribute indication type virtual image information (also referred to as attribute indication information), can be the cruise indication information, or can be a combination of the two. The type and quantity of the virtual image information are not specifically limited in the present application.

[0253] In a possible implementation, the first HUD is a windshield HUD or an augmented reality (AR) HUD. The HUD that can be evaluated by the method provided by the embodiments of the present application at least includes a HUD capable of projecting virtual image information, and the specific type of the first HUD is not limited.

[0254] In a possible implementation, the first image is a key frame in a first video. The first video can be a set of image information obtained by executing a subject to which the present application is applied. Specifically, the obtaining of the first image can include:

[0255] Based on the first video, the first image is obtained. The first video includes a plurality of images, and the first image is an image including virtual image information and real scene information in the plurality of images. It can be understood that the first image can be a single image or a plurality of images. Regardless of whether it is a single image or a plurality of images, the first image includes virtual image information and real scene information.

[0256] It should be noted that the virtual image information and the real scene information included in the single first image are in a corresponding relationship. For example, the virtual image information included in the first image is a lateral identifier indicating a following target, and the corresponding real scene information should be the following target.

[0257] Optionally, in the collected part of the image, there can be virtual image information and real scene information, but the real scene information is not the real scene information corresponding to the virtual image information. Then, this part of the image is an image other than the first image. Illustratively, the method further includes screening of the collected image, the collected image including the first image, and the collected image can be the first video.

[0258] Optionally, the virtual image information and the real scene information included in the first image are one-to-one corresponding.

[0259] Further, it needs to be noted that, in order to verify the fitting degree of the virtual image information projected by the first HUD and the real scene information, and to ensure that the user can obtain correct and accurate cruise instruction information and / or attribute instruction information when driving in various scenes, it is often necessary to test the first HUD device equipped vehicle on actual roads for repeated function testing of multiple road conditions, multiple weathers, multiple regions and multiple scenes, that is, real vehicle testing. In a possible implementation, the first image involved in the embodiments of the present application can be a collection of multiple images obtained in different test scenes, or one or more images obtained in a certain test scene in multiple test scenes. In order to ensure the comprehensiveness of the test scene and the comprehensiveness and accuracy of the fitting degree calculation of the virtual image information in different scenes, a test scene database is constructed in advance, the test scene database includes multiple test scenes, and any two test scenes in the multiple test scenes differ in at least one test factor. It should be understood that any test scene in the multiple test scenes differs from other test scenes in at least one test factor.

[0260] Optionally, the multiple test factors include a road static factor, a dynamic traffic participant factor and an environment factor.

[0261] Optionally, before the step S401, the method further includes:

[0262] The test device constructs a test scene database, the test scene database includes multiple test scenes, any two test scenes in the multiple test scenes differ in at least one test factor, and the multiple test factors include a road static factor, a dynamic traffic participant factor and an environment factor.

[0263] Optionally, the obtaining of the first image can specifically include:

[0264] The test device obtains the first image based on a first scene in the multiple test scenes in the test scene database.

[0265] It can be understood that the multiple test scenes can be obtained by organically combining the multiple test factors in a traversal manner, and different test factors are combined to obtain various test scenes. The first scene can be any scene in the test scene database. This embodiment makes the test process more systematic and standardized, and can cover more possible test scenes and factors. A comprehensive test scene database is constructed.

[0266] In a possible implementation, the multiple test factors further include a test vehicle itself factor, the test vehicle itself factor includes one or more of vehicle basic information, position information or motion state information, and the motion state information includes vehicle speed.

[0267] For example, in part or all of the test scenarios, there are requirements for the speed of the test vehicle on the road, for example, in the same road scene, the speed of the test vehicle should be (40±10) km / h and (80±10) km / h respectively. The main consideration is that the presentation effect of the virtual image information may be different at different speeds, in order to ensure the accuracy of the fitting degree calculation, the effect of the first HUD projected virtual image information carried by the test vehicle in each state (such as each speed) should be considered.

[0268] In a possible implementation, the road static factor includes one or more of intersection shape information, road geometry information, road surface condition information, landmark information, or lane line condition information.

[0269] It should be understood that the intersection shape information can be used to indicate the state of the intersection in the actual road, such as a cross intersection, a fork intersection, etc.; the road geometry information can be used to indicate various parameters of the actual road, such as the slope, the road curvature, the road inclination, etc.; the road surface condition information can be used to indicate the road surface state of the actual road, such as whether there are potholes, whether there are speed reduction zones, or the number or position distribution of speed reduction zones, etc.; the landmark information can be used to indicate the road signs that may appear in the actual road, such as traffic lights, speed limit signs, school ahead signs, etc.; and the lane line condition information can be used to indicate the position of the lane line in the actual road, whether the lane line allows lane changing, and the like.

[0270] In a possible implementation, the dynamic traffic participant factor includes one or more of the relative position relationship between the motor vehicle, the non-motor vehicle, and the pedestrian and the test vehicle. Figure 10 The relative position relationship between the motor vehicle and the test vehicle is exemplarily described with the following follow-up target. Figure 10 In the follow-up target is a motor vehicle, in the case where the follow-up target is far away from the test vehicle, such as part (a) of FIG. 6B, the follow-up target is 10 m away from the test vehicle, and the virtual image information (virtual image 3) corresponding to the follow-up target is a line mark (horizontal line mark), and the specific display position is below the follow-up target. Figure 10 In the follow-up target is a motor vehicle, in the case where the follow-up target is far away from the test vehicle, such as part (a) of FIG. 6B, the follow-up target is 10 m away from the test vehicle, and the virtual image information (virtual image 3) corresponding to the follow-up target is a line mark (horizontal line mark), and the specific display position is below the follow-up target. Figure 11 In the follow-up target is a motor vehicle, in the case where the follow-up target is far away from the test vehicle, such as part (a) of FIG. 6B, the follow-up target is 10 m away from the test vehicle, and the virtual image information (virtual image 3) corresponding to the follow-up target is a line mark (horizontal line mark), and the specific display position is below the follow-up target.

[0271] In one possible implementation, environmental factors include one or more of time, brightness, or weather.

[0272] In the aforementioned implementation method, the test factors are further refined into multiple aspects, including static road factors, dynamic traffic participant factors, environmental factors, and test vehicle-specific factors. This implementation method makes the testing process more comprehensive and detailed, and can more accurately reflect the performance of HUD technology in different scenarios.

[0273] In one possible implementation, the test factor also includes driving function information of the test vehicle, which includes autonomous driving functions or active driving functions configured on the test vehicle, such as ACC function and LCC function. This application does not limit the specific types of driving functions.

[0274] Please see Figure 11 , Figure 11 This is a schematic diagram illustrating an application scenario of a testing method provided in an embodiment of this application. For example... Figure 11 As shown, this application scenario can include computing devices and devices driving on actual roads (e.g., Figure 11 The test vehicle shown is from a real-world multi-lane highway. Figure 11 (Taking passenger cars as an example). Optionally, such as... Figure 11 As shown, this application scenario can also include multiple other vehicles, such as vehicle 1 ( Figure 11 (Taking a passenger car as an example). The computing device and the test vehicle (or the image acquisition device within the test vehicle) can establish a communication connection via wired or wireless means. Figure 11 The diagram illustrates a communication connection established between the computing device and the image acquisition device in the test vehicle via wired or wireless means. Furthermore, the projected image and real-world information of the test vehicle's first HUD are acquired under different testing factors; for example, in the aforementioned road scenario, image acquisition is performed under different vehicle speeds, different weather conditions, and different times of day.

[0275] While driving on this road, the test vehicle uses multiple sensors to collect data about the surrounding environment and sends the data to a computer. Then, as... Figure 12As shown, the computing device can construct a large number of scenes based on the data collected by the plurality of sensors in the test vehicle, which includes the relevant data of the first HUD. Obviously, since the data is collected by driving the real vehicle on the actual road, the large number of scenes can all be real scene simulation scenes. Further, the computing device can construct a virtual HUD based on the preset HUD hardware parameters through software simulation, thereby generating corresponding AR images in the plurality of real scene simulation scenes and fusing them to efficiently and conveniently obtain and display a large number of augmented reality scenes. In this way, the user can intuitively experience the AR function of the HUD product in various scenes without driving the real vehicle on the road, thereby providing convenience for the user. Moreover, the AR function of the HUD can be tested based on the large number of augmented reality scenes.

[0276] Step S402: The test device obtains first features of the virtual image information and second features of the real scene information based on the first image.

[0277] The first features are used to represent the attributes of the virtual image information, and the second features are used to represent the attributes of the real scene information. The attributes include positions and / or directions.

[0278] Optionally, the virtual image information includes cruise instruction information and / or attribute instruction information, and the attribute instruction information includes one or more of pedestrian identification, obstacle identification, front vehicle identification, or lane line identification.

[0279] Optionally, the first features and the second features are for subsequent fitting degree evaluation, and the fitting degree evaluation manners of the virtual image information of different attributes are different.

[0280] Optionally, the first features are determined based on the attributes of the virtual image information, and the second features are determined based on the attributes of the real scene information. It should be understood that different attributes correspond to different features. For example, a triangular mark for guidance and a triangular mark for attribute indication can have the same appearance, but the extracted first features are different.

[0281] The extraction manner and form of the first features of the virtual image information are introduced below.

[0282] The extraction of the first features of the virtual image information can be realized in various manners. For example, the extraction can be realized in the following two manners: Manner one, the test device extracts features based on the appearance of the virtual image information. Manner two, the test device directly extracts features based on the attributes of the virtual image information. The two extraction manners are introduced below.

[0283] Manner one, the extraction of features based on the appearance of the virtual image information.

[0284] The appearance of the virtual image information can be regular or irregular, and the virtual image information can be regular icons and / or irregular icons; it should be understood that the appearance of a single virtual image information, such as an AR icon, can be classified as a regular icon or an irregular icon.

[0285] First, the regular icon is introduced, and the first feature of the regular icon is the geometric center point of the regular icon. The regular icon includes icons that are regular geometric shapes themselves and icons that have regular geometric shapes as outer frames. For details, see Figure 12 , Figure 12 A schematic diagram of a regular icon and the corresponding first feature provided by an embodiment of the present application. In Figure 10 , the regular icon includes line marks (such as horizontal line marks, curved line marks, etc.), and icons with regular geometric shapes as outer frames, such as warning marks with triangular outer frames commonly used for warning.

[0286] Of course, in addition to the above-mentioned line marks and warning marks with triangular outer frames, the regular icon can also include one or more other marks, and the present application does not limit the types and specific shapes of the regular icon. For example, the virtual image 4 in part (b) of Figure 13 also belongs to the regular icon, with a circular outer frame.

[0287] In general, AR icons that do not have a guiding effect generally adopt regular icons or similar appearances to avoid the occurrence of driver distraction in observing the icon, which can lead to safety accidents. Based on this, the first feature extracted from a regular icon that does not have a guiding effect and is fixed in relative position with the real scene information is the geometric center point of the regular icon. Ultimately, the fit degree evaluation for the regular icon is generally based on the relative position of the regular icon and the real scene information, and the shape and appearance of the icon itself are not important, so the geometric center point is used as the first feature of the regular icon.

[0288] Secondly, the irregular icon is introduced, and the first feature of the irregular icon is a key element in the irregular icon, including a key line and / or a key point. The irregular icon generally has a specific function, which is related to its appearance, so the first feature of the irregular icon generally uses important / key boundary, line, point, etc. elements of the irregular icon itself.

[0289] In one possible implementation, the irregular icon includes one or more of a curved mark, an arrow mark, or a road surface cover mark, and the curved mark includes a lane line mark.

[0290] In practical applications, the curvatures of different curves are various, obviously, setting curves with different curvatures is purposeful, for example, indicating lane lines through curve signs, but there are special cases, for example, the lane line itself is straight, but the corresponding curve sign is curved, such cases obviously belong to the case that virtual image and real scene do not match or do not match enough, from here it can be seen that different single identification of the geometric center point of irregular icons as the first feature. The following will describe the common signs that the AR-HUD will project, common signs such as curve signs, arrow signs, and road surface cover signs.

[0291] The following will be described in conjunction with Figure 13 The first feature of the curve sign in FIG. 1 is one or more first key points of the curve sign, and the first key point is the connection point of the tangent line of the curve sign and the curve sign. Figure 13 The first feature of the curve sign in FIG. 1 is four first key points of the curve sign, point 1, point 2, point 3, and point 4, which are all connection points of the tangent line of the curve sign and the curve sign. It should be understood that the number of the first feature of the curve sign is not specifically limited in the present application, and how to evaluate the degree of matching in the case of multiple first features will be described in detail later, which will not be described here.

[0292] In Figure 13 , the first feature of the arrow sign is one or more second key points or first key lines of the arrow sign, and the second key point includes the starting point and the ending point of the arrow sign, such as Figure 13 point 5 and point 6 in FIG. 2, point 5 is the starting point of the arrow sign, and point 6 is the ending point of the arrow sign, and the first key line is a ray from the starting point to the ending point, such as Figure 13 ray 7 in FIG. 2, which is a ray from point 5 through point 6. It should be understood that Figure 13 the specific positions of the starting point and the ending point in FIG. 2 are only for illustration, and the shape and structure of the arrow should also be considered in specific implementation.

[0293] In Figure 13 , the first feature of the road surface cover sign is one or more third key points of the road surface cover sign, and the plurality of third key points includes a center point close to the boundary on one side of the vehicle and a center point away from the boundary on the other side of the vehicle, such as Figure 13 , point 9 is the center point close to the boundary on one side of the vehicle, and point 8 is the center point away from the boundary on the other side of the vehicle. Optionally, the plurality of third key points described above further includes a center point or a tangent point of the boundary on both sides of the road surface cover sign. It can be understood that the center point is generally applied to the case where the boundary is a straight line, and the tangent point is generally applied to the case where the boundary is a curve.

[0294] It should be understood that the arrow mark or the road cover mark in the above-mentioned irregular icon is used to indicate the direction of travel in a general scenario, so two points representing the current position and the estimated position are used as the first feature to represent the specific direction indicated by the icon or the specific pre-arrival position, and the icon playing a guiding / cruising role can only correspond to the first feature representing the corresponding pre-arrival position / pre-travel direction and other guiding indications, as described above Figure 14 The schematic ray 7 can also use a ray from the starting point to the ending point to represent the guiding indication of the corresponding icon.

[0295] The above-mentioned embodiment further refines the first feature extraction method of the irregular icon (such as the curved mark, the arrow mark, the road cover mark, etc.). These feature extraction methods are more in line with the needs of actual application scenarios, making the evaluation results more accurate and reliable.

[0296] Method two, directly extracting features based on the attributes of the virtual image information.

[0297] The above content has described the attributes of the virtual image information in detail, and the attributes of the virtual image information include position and / or direction. It should be understood that the attributes here indicate the attributes common to the virtual image information and the real scene information. For example, the virtual image information includes attribute indication information and cruising indication information. The attribute of the attribute indication information is the position, that is, the position (relative position) of the attribute indication information and the real scene information should be fixed. The attribute of the cruising indication information is the direction, and the corresponding real scene information is generally an intersection, a road, or a lane. The common attribute of the two is the direction, which indicates a direction.

[0298] Obviously, the attribute indication information corresponds to the attribute (position), and the cruising indication information corresponds to the attribute (direction), so method two can also be based on the type of virtual image information for feature extraction.

[0299] It should be understood that the attribute indication information is generally used to indicate the corresponding real scene information (or the name, type, association relationship, etc. of the corresponding real scene information). For example, the attribute indication information includes one or more of the pedestrian mark, the obstacle mark, the front vehicle mark, or the lane line mark. The front vehicle mark is used to indicate the following target, and its general form is a horizontal line below the following target. The obstacle mark is used to indicate the obstacle on the road, and its general form is a triangular warning icon.

[0300] For such icons or attribute indication information, in order to indicate clearly, the relative position with the corresponding real scene is generally fixed, for example, the horizontal line indicating the following target will move with the following target, so the specific shape / structure of such icon does not affect the determination of the fitting degree, and whether the icon is offset from the expected position can be directly used to judge whether it is fitted, so the geometric center point of such icon can be directly used as the first feature, which is simple to extract, clear to indicate, and can reduce resource consumption during evaluation.

[0301] For cruise indication information, generally, the cruise indication information is indicated by arrow identification or road surface cover identification, the role of such icon is cruise / guidance, a single point cannot represent its role, so the corresponding feature can be indicated by key element. In a possible implementation, the virtual image information includes cruise indication information, the first feature of the cruise indication information is a fourth key point or a second key line of the cruise indication information, the fourth key point includes a start point and an end point of the cruise indication information, and the second key line is a ray from the start point to the end point.

[0302] The role of the above-mentioned cruise indication information is cruise / guidance, a single point cannot represent its role, so the start point and the end point representing the direction, or the ray from the start point to the end point can be used as the first feature point, which is simple to extract, clear to indicate, and can reduce resource consumption during evaluation.

[0303] Of course, other ways can also be used for feature extraction in the specific implementation process, for example, using an artificial intelligence model for extraction, and the present application is also applicable to these extraction methods.

[0304] In a possible implementation, the second feature of the real scene information is determined based on the target projection position of the virtual image information, and the target projection position is the position projected by the virtual image information. Optionally, in different functional scenarios, the target projection position of the virtual image information is different. Optionally, the second feature of the real scene information is related to the attribute of the real scene information. Taking the front vehicle identification as an example, the target projection position of the front vehicle identification is the lower edge of the outer contour of the front vehicle, but the front vehicle identification is specifically represented by a horizontal line, and the corresponding first feature is a geometric center point, so the second feature of the real scene information (the front vehicle) is the geometric center point of the target projection position (the lower edge of the outer contour of the front vehicle).

[0305] Step S403: The test device determines the fitting degree evaluation result of the first HUD based on the first feature and the second feature.

[0306] The fitting degree evaluation result is used to indicate the fitting degree of the projection image of the first HUD and the real driving environment.

[0307] In a possible implementation, the fitting degree evaluation result is obtained based on a plurality of fitting degrees, the plurality of fitting degrees including fitting degrees of different virtual image information and corresponding real scene information. Alternatively, the plurality of fitting degrees further include fitting degrees of the same virtual image information and corresponding real scene information in different scenes.

[0308] In summary, in some embodiments, the fitting degree evaluation result needs to be obtained through multiple fitting degree evaluations of the first feature and the second feature. Specifically, based on the first feature and the second feature, the fitting degree evaluation result of the first HUD is determined, including: determining a first fitting degree of the first feature and the second feature; and determining the fitting degree evaluation result of the first HUD based on the first fitting degree. It should be understood that the first feature and the second feature are collectively referred to, which can indicate the features of a virtual image information and corresponding real scene information, or the features of a plurality of virtual image information and corresponding real scene information. Further, the first fitting degree can be understood as a quantitative or qualitative evaluation of the fitting degree of a virtual image information and corresponding real scene information, or a quantitative or qualitative evaluation set of a plurality of virtual image information and corresponding real scene information.

[0309] Embodiments of the present application propose quantitative and qualitative evaluation methods for the fitting degree evaluation of the HUD, which are described below.

[0310] 1. Quantitative evaluation.

[0311] In the embodiments of the present application, for the fitting degree evaluation between virtual image information and real scene information with fixed relative positional relationship, a quantitative evaluation method can be used. Exemplarily, the quantitative evaluation is an evaluation method based on specific numerical values or measurement standards. The evaluation process using this evaluation method can involve the calculation of some formulas. The evaluation result obtained by the quantitative evaluation is generally a specific numerical value or a numerical range, for example, degrees, meters. For example, the result obtained by the quantitative evaluation can be that the angle between the real scene information and the virtual image information is x degrees, or the distance between the real scene information and the virtual image information is m meters, and other evaluation results involving specific numerical values. Then, the deviation degree or positional relationship between the virtual image information and the real scene information is accurately described through specific numerical values / numerical ranges.

[0312] Alternatively, the quantitative evaluation is a method of obtaining an evaluation result based on some parameters of the real scene information and the virtual image information. Exemplarily, the parameters can be the angle between the two, the distance between the two, and other parameters. The evaluation result is obtained based on the above parameters, which is generally represented in the form of numerical values, such as percentages. For example, the fitting degree evaluation result of a certain real scene information and a certain virtual image information is 89%.

[0313] It should be noted that the manifestation form and specific value of the evaluation result involved in the embodiments of the present application are only examples, and the present application does not limit this.

[0314] In a possible implementation, the first fitting degree is related to the positions of the first feature and the second feature, in other words, the determination manner of the first fitting degree is related to the positions of the first feature and the second feature. Optionally, the determination manner of the first fitting degree includes determination based on the relative positional relationship of the first feature and the second feature. Exemplarily, the relative positional relationship includes the distance between the first feature and the second feature.

[0315] Considering that the heights and postures of different drivers are different, the relative positional relationship of the first feature and the second feature may be different under the visual angle of different drivers. Then, the embodiments of the present application consider that the first feature and the second feature have more positional relationship in a three-dimensional space. A single distance is difficult to serve as a quantitative evaluation standard in some scenarios.

[0316] Based on the above description, the relative positional relationship between the first feature corresponding to the attribute indication information and the second feature corresponding to the real scene information should be fixed. Based on this, the fitting degree of the attribute indication information and the corresponding real scene information can be evaluated.

[0317] Next, taking the attribute indication information as an example, a quantitative evaluation manner of the fitting degree between the virtual image information and the real scene information is introduced. In combination with Figure 14 Exemplarily, Figure 14 A schematic diagram of the fitting degree provided by the embodiments of the present application.

[0318] In a possible implementation, the above-mentioned determination of the first fitting degree of the first feature and the second feature, especially in the case that the first fitting degree is determined based on the positions of the first feature and the second feature, can include:

[0319] Based on the first feature, the second feature and the starting point of the driver's visual angle, the first fitting degree between the first feature and the second feature is determined. The first fitting degree is used to measure the fitting deviation between the real scene information and the virtual image information. The first fitting degree is a value based on the included angle formed by the first feature and the second feature compared with the starting point of the driver's visual angle. Then, based on the first fitting degree, the first fitting degree of the first feature and the second feature is determined.

[0320] In combination with Figure 14The virtual image information is used as the front vehicle identifier, and the second feature of the real scene information is determined by the target projection position of the virtual image information. The front vehicle identifier is attribute indication information, and the relative position relationship with the real scene information (the front vehicle) is fixed. The front vehicle identifier should be projected at the center of the lower edge of the contour of the front vehicle. In combination with the description of the first feature of the attribute indication information extracted according to the first feature, the first feature of the front vehicle identifier is the geometric center point of the corresponding identifier, as shown by point A in FIG. 6A. Point A is the first feature of the corresponding identifier (virtual image information) of the following target in the driver's perspective, that is, the horizontal center of the following target corresponding identifier (the front vehicle identifier in the virtual image information) recognized in the driver's perspective. Figure 14

[0321] Correspondingly, the second feature of the real scene information is determined by the target projection position of the virtual image information. The target projection position of the virtual image information is the lower edge of the contour of the front vehicle in the driver's perspective, and is projected at the center. The second feature of the real scene information is the geometric center point of the target projection position, as shown by point B in FIG. 6B. Point B is the center point of the lower edge of the contour of the front vehicle in the driver's perspective. Figure 15

[0322] Point O is the center point of the eyebox, which is the starting point of the driver's perspective, that is, the eye point mentioned above. Details are not described here.

[0323] The fitting degree α is ∠AOB.

[0324] In the above-mentioned embodiments, a method for determining the first fitting degree based on the first feature, the second feature, and the starting point of the driver's perspective is proposed. This method measures the fitting deviation between the real scene information and the virtual image information by calculating the included angle, thereby realizing the quantitative evaluation of the fitting degree.

[0325] It should be understood that the target projection position corresponding to different virtual image information may be different. For example, when the test vehicle is close to the front vehicle, the front vehicle identifier becomes a warning identifier, and the target projection position of the warning identifier is different from the target projection position of the front vehicle identifier mentioned above, and may be the geometric center of the contour of the front vehicle in the driver's perspective.

[0326] ​​But in some scenarios, the evaluation method of the fitting degree is different from the above method. For example, in a scenario where the virtual image information is a lane marking, generally, the lane marking is used to prompt the driver of the lane line on the road, for example, prompting the driver that there is a vehicle approaching behind the lane line, and not to drive beyond the lane line. Therefore, the lane marking is generally coincident with the actual lane line. It can be understood that the lane marking is a special case of the virtual image information. Specifically, the relative position between the lane marking and the real scene information should be fixed. Therefore, based on the above description, the corresponding first feature should be the geometric center point of the marking. However, in actual application, the lane marking may be partially coincident with the actual lane line or partially non-coincident. For example, the lane marking itself is a curve. The curved lane marking can be applied to a scenario where the actual road lane line itself is a curve, or a scenario where the actual road lane line is a straight line, but the projected lane marking is a curve. Therefore, it is obviously unreasonable to directly identify the first feature as the geometric center point of the lane marking. Therefore, there are multiple first key points of the first feature of the lane marking. For such a situation, in one possible implementation, referring to Figure 15 , Figure 15 , a second fitting degree diagram provided by an embodiment of the present application is shown in Figure 15 , the virtual image information is a lane marking, and the corresponding first feature is multiple first key points of the lane marking. The first fitting degree between the first feature and the second feature based on the first feature, the second feature, and the starting point of the driver's perspective is determined, including:

[0327] The second fitting degree between the first feature and the second feature based on the first feature, the second feature, and the starting point of the driver's perspective is determined, and the second fitting degree includes multiple second fitting degrees. The third fitting degree with the maximum value in the multiple second fitting degrees is determined as the first fitting degree. As shown in Figure 15 , the points 10, 11, and 12 are the first feature, and the corresponding second features are the intersection of the normal line of the tangent line at the point 10 and the lane line, the intersection of the normal line of the tangent line at the point 11 and the lane line, and the intersection of the normal line of the tangent line at the point 12 and the lane line. Correspondingly, the second fitting degrees can be the angle between the line segment OD and the line segment O(point 10), the angle between the line segment OE and the line segment O(point 11), and the angle between the line segment OF and the line segment O(point 12). Obviously, the fitting deviation of the point D and the point 10 is larger, and the corresponding fitting angle is also larger. Therefore, the third fitting degree Max a is determined.

[0328] It should be noted that the above Figure 15 diagram is only used to explain the second fitting degree and the third fitting degree. In actual application, the second fitting degree can include a large number of points on the lane marking, for example, each point of curvature change is selected as a first feature point, and then the second feature point is determined, and finally the second fitting degree is determined.

[0329] The third fitting degree with the largest value in the plurality of second fitting degrees is determined as the first fitting degree in the case where there are a plurality of second fitting degrees, and the third fitting degree represents that the corresponding first key point / first feature is farthest away from the second feature in the three-dimensional space, and the fitting deviation is the largest, such as Maxa in the following formula: Figure 16 The first fitting degree of the lane line mark identified by the third fitting degree meets the accuracy requirement of the fitting degree evaluation.

[0330] In the above-mentioned embodiments, a specific method for determining the first fitting degree is proposed for the case where the virtual image information is a lane line mark. By calculating the maximum value in the plurality of second fitting degrees as the first fitting degree, the evaluation result is more in line with the requirements of the actual application scenario. This embodiment helps to improve the accuracy and reliability of the HUD technology in lane line marking.

[0331] In a possible implementation, the greater the value of the first fitting degree, the lower the fitting degree.

[0332] In a possible implementation, the first fitting degree includes a first horizontal fitting degree and a first vertical fitting degree. As the name implies, the horizontal fitting degree is the angle between the first feature and the second feature in the horizontal direction, and the vertical fitting degree is the angle between the first feature and the second feature in the vertical direction. The horizontal fitting degree and the vertical fitting degree are generally used to evaluate virtual image information with a line display effect. Generally, both the first horizontal fitting degree and the first vertical fitting degree meet the requirements, or both are 0.

[0333] The possible implementation of the method for determining the first fitting degree is specifically given below. In a possible implementation, the first fitting degree between the first feature and the second feature is determined based on the first feature, the second feature, and the starting point of the driver's perspective, and includes:

[0334] A first coordinate system and a second coordinate system are established. The first coordinate system is related to the first image, and the origin of the first coordinate system is the top left corner vertex of the image. The second coordinate system is a three-dimensional coordinate system, and the origin of the second coordinate system is the starting point of the driver's field of view. The first image is obtained by an image acquisition device, and the center point of the lens position of the image acquisition device is the starting point of the driver's field of view. It should be understood that during the test process, the image acquisition device can continuously acquire, and accordingly, the first coordinate system and the second coordinate system can be partially implemented in advance, such as before step S402.

[0335] The testing device determines a first coordinate of the first feature in the first coordinate system and a second coordinate of the second feature in the first coordinate system; the testing device converts the first coordinate and the second coordinate into a third coordinate and a fourth coordinate respectively based on intrinsic information and extrinsic information of the image acquisition device, and the first coordinate system and the second coordinate system, the third coordinate being a coordinate of the first feature in the second coordinate system, and the fourth coordinate being a coordinate of the second feature in the second coordinate system; and the testing device determines the first fit degree between the first feature and the second feature based on the third coordinate and the fourth coordinate. Please refer to Figure 17 and Figure 16 , Figure 17 A first coordinate system provided by an embodiment of the present application is shown in the following figure, Figure 16 A second coordinate system provided by an embodiment of the present application is shown in the following figure. Figure 14 The origin of the first coordinate system in the above-mentioned figure is the top-left corner of the first image, and the second coordinate system is a three-dimensional coordinate system, and the origin of the second coordinate system is the starting point of the driver's field of view.

[0336] In an alternative embodiment, a coordinate conversion formula is provided, which is as follows:

[0337]

[0338] wherein K1 is the camera intrinsic and K2 is the camera extrinsic, the first coordinate or the second coordinate in the first coordinate system is (u, v), and the third coordinate and the fourth coordinate in the second coordinate system are (X, Y, Z).

[0339] It should be noted that the calibration parameters can be determined in various ways, for example, according to Zhang Zhengyou's calibration method.

[0340] Considering that the image acquisition device is placed at the eye point position, the extrinsic matrix K2 can be ignored.

[0341] Alternatively, the first coordinate and the second coordinate can be obtained by comparing the pixels of the first feature and the second feature, and then the third coordinate and the fourth coordinate can be obtained by the above formula.

[0342] Alternatively, the fit degree can be obtained by vector dot product, for example, based on the third coordinate and the fourth coordinate, a first vector representing the first feature and a second vector representing the second feature are obtained, and then the first vector and the second vector are calculated by inner product, thereby obtaining the fit degree, which can also be called the fit angle.

[0343] In the above-mentioned embodiments, the specific method of determining the first fit degree based on the first feature, the second feature and the starting point of the driver's perspective is described in detail. By establishing the first coordinate system and the second coordinate system, and using the intrinsic information and extrinsic information of the image acquisition device for coordinate conversion, the accurate description of the first feature and the second feature in the three-dimensional space is realized.

[0344] It is understandable that the starting point of the driver's perspective mentioned above specifically refers to the driver's eye point. The first image includes the driver's field of vision. Therefore, the above-mentioned fitting calculation converts the relative positional relationship between the first feature and the second feature into an angular relationship in three-dimensional space. Considering that different drivers have different sitting postures and heights, the positions between the first feature and the second feature directly observed by the driver are different on the two-dimensional plane. Therefore, using the positional relationship (angular relationship) in three-dimensional space can more accurately and objectively measure the fitting deviation between real scene information and virtual image information.

[0345] Of course, in addition to the above-mentioned method for calculating the fit, the degree of fit involved in the embodiments of this application can also be calculated by other quantitative methods. For example, a formula for calculating the fit is provided, where the fit can also be called the fit deviation β, as follows:

[0346]

[0347] Where m represents system enhanced display information (virtual image information, such as...) Figure 14 The linear dimension of the fit deviation between the vehicle identification mark in the middle (the vehicle in front) and the corresponding real scene (the lower edge of the outer contour of the vehicle in front);

[0348] a eye To enhance the display information of the system (virtual image information, such as...) Figure 18 The virtual image distance (where the vehicle in front is identified) can be the distance between the projected virtual image and the human eye's pupil (VID). This data can be obtained by testing the optical parameters of the image acquisition device and / or the first HUD before the test begins using the aforementioned optical testing module, or it can be obtained based on the product data of the test vehicle.

[0349] 2. Qualitative assessment.

[0350] Considering that the relative positional relationship between the cruise guidance information and the corresponding real-world information is not fixed, a qualitative approach can be used to assess the degree of fit of the cruise guidance information.

[0351] Exemplarily, the qualitative evaluation is an evaluation manner based on a descriptive and judgmental standard. In a partial detection scene, the specific positions of the virtual image information are not anchored to a certain real scene information or a certain scene, and it is difficult to accurately describe the fitting degree thereof by specific numerical values, and therefore a standard more in line with the experience and driving habits of the driver is used to define the fitting degree. The evaluation result obtained by the qualitative evaluation is usually presented in a descriptive manner, for example, "fitting", "not fitting", "high fitting degree", "medium fitting degree", etc., or in a grading manner, such as "first level", "second level", "third level", etc., or in a road grading manner, such as "intersection level fitting", "road level fitting", "lane level fitting", etc., for indicating the fitting degree evaluation result capable of realizing an intersection indication effect, a road indication effect or a lane indication effect.

[0352] It should be noted that the expression form of the evaluation result described above is only an example, and the present application is not limited thereto.

[0353] In a possible implementation, the first fitting degree is also related to the type of the virtual image information, and the type of the virtual image information includes a cruise indication type. In other words, the determination manners of the first fitting degree corresponding to different types of virtual image information are different. In the present embodiment, the determination manner of the fitting degree corresponding to the virtual image information of the cruise indication type is different from that of the virtual image information of other types.

[0354] By considering the specific requirements of the virtual image information in actual application (such as the accuracy and continuity of the cruise indication information), the evaluation result is more in line with the requirements of the actual application scene. In a possible implementation, the first fitting degree is determined based on a target condition satisfied by the virtual image information.

[0355] Generally, the cruise indication information is generally indicated by arrow identification or road surface overlay identification. First, the qualitative fitting degree evaluation manner is exemplarily introduced by the target condition of the arrow identification.

[0356] In combination Figure 18 In a possible implementation, the virtual image information is cruise indication information, and the cruise indication information includes arrow identification. In the case where the cruise indication information is arrow identification, in order to ensure the cruise / guiding effect of the arrow identification, the corresponding target condition is set as a condition related to the cruise / guiding effect. Exemplarily, the target condition includes at least one of the following:

[0357] The cutoff point of the arrow identification is located in a target road, and the target road includes a target lane;

[0358] The cutoff point of the arrow identification is located in the target lane;

[0359] The extension line extending in the direction indicated by the arrowhead marker intersects the target road, and the distance between the intersection point and the vehicle is less than the first distance;

[0360] The extension line extending in the direction indicated by the arrowhead marker intersects the target lane, and the distance between the intersection point and the vehicle is less than the second distance.

[0361] It should be understood that the target road is the road to be traveled estimated by the test vehicle during cruising, the target lane is the lane to be traveled estimated by the test vehicle during cruising, the first feature of the test arrowhead marker is one or more second key points of the arrowhead marker, or the first key line, and the corresponding real scene information is the target road or the target lane.

[0362] In order to facilitate understanding of the above target conditions, please refer to Figure 18 , Figure 18 The above target conditions are met by mainly giving various possible implementations of the arrowhead marker.

[0363] In the (a) part of Figure 18 , the end point of the arrowhead marker (the second key point in the first feature of the arrowhead marker) does not fall within the target road or the target lane, but the first key line in the first feature of the arrowhead marker intersects the target road, and the distance between the intersection point and the vehicle is less than the first distance, which meets the target condition that the extension line extending in the direction indicated by the arrowhead marker intersects the target road, and the distance between the intersection point and the vehicle is less than the first distance (not shown in Figure 18 ). It should be understood that the first distance can be 10m, which is a distance that can be observed by the driver in the vehicle in most scenarios, of course, 10m here is only an example, the key point is that the first distance should not exceed the driver's sight distance, or be too far from the test vehicle, and the specific data used in different implementation scenarios is different. It should be noted that the first key line intersects the target lane, but the distance between the intersection point and the vehicle is greater than the second distance (not shown in Figure 18 ).

[0364] In the (b) part of Figure 18 , the end point of the arrowhead marker still does not fall within the target road or the target lane, but the first key line in the first feature of the arrowhead marker intersects the target road and the distance between the intersection point and the vehicle is less than the first distance, and the first key line intersects the target lane and the distance between the intersection point and the vehicle is less than the second distance (not shown in Figure 18 ). This represents that the arrowhead marker in the (b) part meets the target conditions of "the extension line extending in the direction indicated by the arrowhead marker intersects the target road, and the distance between the intersection point and the vehicle is less than the first distance, and the extension line extending in the direction indicated by the arrowhead marker intersects the target lane, and the distance between the intersection point and the vehicle is less than the second distance".

[0365] The second distance is set away from the first distance in a similar principle, but in some implementation scenarios, the first distance is different from the second distance, and the second distance is smaller than the first distance.

[0366] It should be understood that, Figure 18 The first distance and the second distance are not shown in the above embodiments, and in general, the first distance and the second distance are distances between the intersection points and the test vehicle.

[0367] In the (c) part of the above embodiments, Figure 18 In the (c) part of the above embodiments, the arrow-identified end point falls within the target road but does not fall within the target lane, satisfying the “arrow-identified end point is located within the target road” in the target condition. It should be understood that if an arrow identification satisfies the “arrow-identified end point is located within the target road” in the target condition, the corresponding also satisfies the “the extension line extending in the direction indicated by the arrow identification intersects with the target road, and the distance between the intersection point and the vehicle is less than the first distance” in the target condition, and it is also possible to satisfy the “the extension line extending in the direction indicated by the arrow identification intersects with the target lane, and the distance between the intersection point and the vehicle is less than the second distance” in the target condition.

[0368] In the (d) part of the above embodiments, Figure 19 In the (d) part of the above embodiments, the arrow-identified end point falls within the target lane, and compared with other parts of the arrow identification, the guidance accuracy is the highest, and in addition to satisfying the “arrow-identified end point is located within the target lane” in the above-mentioned target condition, it also generally satisfies the other three target conditions.

[0369] In the above-mentioned embodiments, optionally, the target condition includes a condition for indicating the direction of the vehicle from the actual feeling of the driver. It can be understood that the target condition includes a plurality of conditions corresponding to different levels, such as the end point being located in the target road and the end point being located in the target lane. It can be obviously seen that the guidance effect of the end point being located in the target lane is stronger than that of the end point being located in the target road. Optionally, the target condition corresponding to the cruise indication information should also include the condition of no misleading.

[0370] It should be noted that the qualitative evaluation can be based on the fitting degree evaluation of the arrow mark meeting the target condition. For example, the fitting degree of the arrow mark can be divided into "intersection level", "road level" and "lane level" in order from low to high. The "arrow mark end point in the target lane" in the target condition belongs to the lane level, the "arrow mark end point in the target road" in the target condition belongs to at least the road level, and the "extension line extending along the direction indicated by the arrow mark intersects the target road, and the intersection distance between the intersection point and the vehicle is less than the first distance; the extension line extending along the direction indicated by the arrow mark intersects the target lane, and the intersection distance between the intersection point and the vehicle is less than the second distance" in the target condition belongs to at least the intersection level. Of course, in addition to this, the fitting degree can be further divided into more levels, which is not limited in the present application.

[0371] In a possible implementation, the first fitting degree is determined based on the target condition meeting the virtual image information and the first scene, and the first scene belongs to the scene supported by the first HUD.

[0372] In a possible implementation, the virtual image information is cruise indication information, the cruise indication information includes an arrow mark, the first scene includes a target road curvature greater than 0, and the target condition further includes consistency (or close consistency) with the road curve tendency of the target road. The target road is the road to be traveled by the estimated vehicle. In a specific scene, such as the above-mentioned target road curvature greater than 0 or the scene of the fork intersection, the cruise indication should clearly indicate the corresponding road and cannot be misleading. Except that the current lane of the vehicle is a straight forward lane (the curvature is approximately equal to 0), but the target road is a fork intersection, which may have a curvature, i.e., the curvature of the target road is greater than 0, then the cruise indication information only needs to indicate the position of the target road, or the tendency is consistent with the road curve of the target road, even if the end point does not fall within the target road / lane, it still plays a guiding role, therefore, in such a scene, the target condition further includes consistency with the road curve tendency of the target road.

[0373] Optionally, the "consistency with the road curve tendency of the target road" described above can be replaced by "equal to or close to the curvature of the target road".

[0374] It should be understood that the above description of the target condition, options and grading of the fitting degree are only examples, which are not limited in the present application.

[0375] Next, the qualitative fitting degree evaluation method is further described by the target condition of the road surface cover mark.

[0376] Please refer to Figure 19 , Figure 19Fig. 1 is a schematic diagram of a plurality of possible implementations of a road surface overlay marking according to an embodiment of the present application. In one possible implementation, the virtual image information is cruise instruction information, the cruise instruction information includes a road surface overlay marking, and the corresponding target condition includes at least one of the following:

[0377] The center point of the boundary near the side of the vehicle (test vehicle) is located in front of the vehicle (test vehicle);

[0378] The center point of the boundary far from the side of the vehicle (test vehicle) is located in a target road, which is a road that the vehicle (test vehicle) is estimated to travel on, and the target road includes a target lane;

[0379] The center point of the boundary far from the side of the vehicle (test vehicle) is located in the target lane;

[0380] The road surface overlay marking has a natural shape;

[0381] The road surface overlay marking is continuous and changes smoothly.

[0382] The natural shape of the road surface overlay marking and the continuous and smooth change of the road surface overlay marking are mainly used to indicate the shape of the road surface overlay marking, and the required standard is natural, without singular points.

[0383] The first feature of the road surface overlay marking is one or more third key points of the road surface overlay marking. If there is only one third key point, it can be the center point of the boundary near the side of the vehicle or the center point of the boundary far from the side of the vehicle. If there are multiple third key points including the center point of the boundary near the side of the vehicle and the center point of the boundary far from the side of the vehicle, the present embodiment is described by taking multiple third key points as an example.

[0384] For ease of understanding, please refer to Figure 19 (a) part, the road surface overlay marking in the (a) part meets the target condition of "the center point of the boundary near the side of the test vehicle (third key point) is located in front of the vehicle (test vehicle), and the center point of the boundary far from the side of the vehicle (test vehicle) is located in the target road; the center point of the boundary far from the side of the vehicle (test vehicle) is located in the target lane".

[0385] Please refer to Figure 19 (b) part, the road surface overlay marking in the (b) part meets the target condition of "the center point of the boundary near the side of the vehicle (test vehicle) is located in front of the vehicle (test vehicle); the center point of the boundary far from the side of the vehicle (test vehicle) is located in the target road".

[0386] It can be understood that the road surface overlay marking mainly plays a guiding and indicating role. In terms of the fitting degree evaluation, the more accurate the guiding and indicating is, the higher the fitting degree is. Obviously, the above Figure 19the center point of the boundary of the road surface cover identification on the side away from the vehicle in the (a) part is located in the target lane, and Figure 19 the center point of the boundary of the road surface cover identification on the side away from the vehicle in the (b) part is located in the target road, and not in the target lane. Figure 19 the road surface cover identification in the (a) part should have a higher degree of fitting than Figure 19 the road surface cover identification in the (b) part.

[0387] Exemplarily, the degree of fitting of the road surface cover identification is divided into “road level” and “lane level”, and the “lane level” is higher than the “road level”, and the above Figure 19 the road surface cover identification in the (a) part meets the “lane level” determination requirement, and the above Figure 19 the road surface cover identification in the (b) part meets the “road level” determination requirement. In addition, the degree of fitting can also be divided into a level of “not meeting the fitting requirement” or similar naming to correspond to the road surface cover identification that does not meet the above target conditions.

[0388] Optionally, the target conditions of the road surface cover identification can also include one or more of the following: the guide line trend is consistent with the planned road level trend, the guide line trend is consistent with the planned lane level trend, the guide line is the trend of the road surface cover identification, such as the boundary line or part of the boundary line on both sides of the road surface cover identification can be used as a guide, and the road level trend and the lane level trend can be respectively referred to Figure 19 the trend of the road surface cover identification in the (a) part, and Figure 19 the trend of the road surface cover identification in the (b) part.

[0389] In the above-mentioned embodiments, the first degree of fitting is evaluated in combination with the scene supported by the HUD. This embodiment makes the evaluation process more specific and targeted, and can be customized for the needs of different scenes. The introduction of this step helps to improve the adaptability and accuracy of the HUD evaluation method in different scenes.

[0390] Considering the case that a certain identification may meet multiple target conditions, in one possible embodiment, the first degree of fitting is determined based on the number of target conditions met by the virtual image information, and the more target conditions met by the virtual image information, the higher the degree of fitting.

[0391] Based on the above qualitative and quantitative evaluation methods for evaluating the degree of fitting, the first degree of fitting of the first feature and the second feature in the first image can be obtained.

[0392] In a possible implementation, the first image includes a second image and a third image, the second image and the third image are images spaced by a first time in the first image, the first fitting degree includes a second fitting degree and a third fitting degree, the second fitting degree is a fitting degree of virtual image information and real scene information in the second image, and the third fitting degree is a fitting degree of virtual image information and real scene information in the third image. It can be understood that the above-mentioned interval of the first time can be 0 seconds or n seconds, and the present application does not limit this.

[0393] Based on this, based on the first fitting degree, the fitting degree evaluation result of the first HUD is determined, including:

[0394] Based on the second fitting degree and the third fitting degree, the fitting degree evaluation result of the first HUD is determined, and the fitting degree evaluation result includes a jitter evaluation result.

[0395] Please refer to Figure 20 The jitter evaluation result includes a jitter amplitude and a jitter frequency, the jitter amplitude is used to indicate a maximum fluctuation degree of the fitting degree corresponding to the first image, and the jitter amplitude includes a single-period jitter amplitude and a jitter distribution dispersion degree. The single-period jitter amplitude includes a first jitter amplitude obtained based on the second fitting degree and the third fitting degree, and details can be referred to Figure 20 In the first jitter amplitude, α1 and α2 correspond to a time interval of the first time, and the second fitting degree and the third fitting degree are the maximum value and the minimum value in the first time, respectively. It should be understood that the time intervals are different in different periods, and in the process of determining the single-period jitter amplitude, the time scale is used as the division to determine the maximum fitting degree and the minimum fitting degree in each continuous time period.

[0396] The single-period jitter amplitude includes a first jitter amplitude obtained by subtracting the second fitting degree from the third fitting degree, and the first jitter amplitude is α1 minus α2 in Figure 20 For example, the first jitter amplitude in the corresponding period is α1 minus α2.

[0397] Optionally, the time scales of the continuous time periods can be the same or different.

[0398] The embodiment of the present application provides a test method, which obtains an image under the driver's perspective of a test vehicle, the image containing real scene information and virtual image information generated by a first HUD. By extracting the features of the two kinds of information, and determining the fitting degree evaluation result of the HUD based on these features, the objective evaluation of the fitting degree of the HUD projection image and the real driving environment is realized. Compared with the traditional subjective judgment, this method provides more accurate and quantitative evaluation criteria, which helps to improve the performance and user experience of the HUD technology.

[0399] The method of the embodiments of the present application is introduced above, and some apparatuses for implementing the foregoing method are introduced below. It should be understood that the apparatuses provided in the embodiments of the present application are only logical function division, and all or part of the apparatuses can be integrated into one physical entity, or can be physically separated. In addition, the units in the apparatuses can be implemented in the form of processor calling software; for example, the apparatus includes a processor, the processor is connected with a memory, the memory stores instructions, and the processor calls the instructions stored in the memory to implement any one of the above methods or to implement the functions of the units of the apparatus, wherein the processor is, for example, a general processor such as a central processing unit (CPU) or a microprocessor, and the memory is an internal memory of the apparatus or an external memory of the apparatus. Alternatively, the units in the apparatus can be implemented in the form of hardware circuit, and the functions of part or all of the units can be implemented through the design of the hardware circuit, which can be understood as one or more processors; for example, in one implementation, the hardware circuit is an application-specific integrated circuit (ASIC), and the functions of part or all of the units are implemented through the design of the logical relationship of elements in the circuit; for example, in another implementation, the hardware circuit is a programmable logic device (PLD), and taking a field programmable gate array (FPGA) as an example, the FPGA can include a large number of logic gate circuits, and the connection relationship between the logic gate circuits is configured through a configuration file, so as to implement the functions of part or all of the units. All the units of the above apparatuses can be implemented in the form of processor calling software, or all the units can be implemented in the form of hardware circuit, or part of the units are implemented in the form of processor calling software, and the remaining part is implemented in the form of hardware circuit.

[0400] In embodiments of the application, the processor is a circuit having a processing capability of signals. In one implementation, the processor can be a circuit having a capability of reading and running instructions, such as a central processing unit (CPU), a microprocessor, a graphics processing unit (GPU) (which can be understood as a microprocessor), a digital signal processor (DSP), etc. In another implementation, the processor can implement certain functions through a logical relationship of a hardware circuit, which is fixed or reconfigurable. For example, the processor is a hardware circuit implemented by an application-specific integrated circuit (ASIC) or a programmable logic device (PLD), such as an FPGA. In the reconfigurable hardware circuit, the processor loads a configuration document to implement the configuration of the hardware circuit. It can be understood that the processor loads instructions to implement the functions of the above units. In addition, it can also be a hardware circuit designed for artificial intelligence, which can be understood as an ASIC, such as a neural network processing unit (NPU), a tensor processing unit (TPU), a deep learning processing unit (DPU), etc.

[0401] It can be seen that each unit in the above apparatus can be one or more processors (or processing circuits) configured to implement the above methods, for example: CPU, GPU, NPU, TPU, DPU, microprocessor, DSP, ASIC, FPGA, or a combination of at least two of these processor forms.

[0402] In addition, each unit in the above apparatus can be integrated together or can be independently implemented. In one implementation, the units are integrated together to implement a system-on-a-chip (SOC). The SOC can include at least one processor for implementing any of the above methods or functions of the units of the apparatus. The at least one processor can be different, such as a CPU and an FPGA, a CPU and an artificial intelligence processor, a CPU and a GPU, etc.

[0403] The following are several possible apparatuses.

[0404] See Figure 21 , Figure 21This is a schematic diagram of a testing device provided in an embodiment of this application. Optionally, the testing device 210 can be a standalone device, such as a personal computer. Alternatively, the testing device 210 can also be a component within a standalone device (such as a node), such as a chip or integrated circuit. The testing device 210 is used to implement the method executed by the testing device in the aforementioned testing method, for example... Figure 4 In the embodiments shown, the method executed by the testing device in any one or more of the test methods shown in the embodiments.

[0405] like Figure 21 As shown, the testing device 210 includes a communication unit 2101 and a processing unit 2102. The communication unit 2101 is used to perform one or more operations such as acquiring, receiving, listening, transmitting, establishing a connection, and responding, and further includes other operations for implementing the testing method. The processing unit 2102 is used to perform one or more operations such as processing, calculating, determining, and generating, and further includes other operations for implementing the testing method.

[0406] For related descriptions, please refer to Figure 4 The embodiments shown are described in detail here.

[0407] Please see Figure 22 , Figure 22 This is a schematic diagram of a computing device provided in an embodiment of this application. The computing device 220 can be a standalone device, such as a node, or a component included within a standalone device, such as a chip, software module, or integrated circuit. The computing device 220 may include at least one processor 2201 and a communication interface 2202. Optionally, it may also include at least one memory 2203. Further optionally, it may include a connection line 2204, wherein the processor 2201, the communication interface 2202, and / or the memory 2203 are connected via the connection line 2204, and / or communicate with each other via the connection line 2204 to transmit control signals and / or data signals.

[0408] in:

[0409] The processor 2201 is a module for performing arithmetic operations and / or logical operations, and can specifically include one or more of the following modules: a filter, a modem, a power amplifier, a low noise amplifier (LNA), a baseband processor, a radio frequency processor, a radio frequency circuit, a central processing unit (CPU), an application processor (AP), a microcontroller unit (MCU), an electronic control unit (ECU), a graphics processing unit (GPU), a microprocessor unit (MPU), an application specific integrated circuit (ASIC), an image signal processor (ISP), a digital signal processor (DSP), a field programmable gate array (FPGA), a complex programmable logic device (CPLD), or a co-processor, etc.

[0410] The communication interface 2202 can be configured to provide information input or output for the at least one processor, or to receive a signal transmitted from an external device and / or transmit a signal to an external device.

[0411] For example, the communication interface 2202 can include an interface circuit.

[0412] For example, the communication interface 2202 can include a wired link interface such as an Ethernet cable, and can also be a wireless link (Wi-Fi, Bluetooth, universal wireless transmission, vehicle-mounted short-range communication technology, and other short-range wireless communication technologies) interface.

[0413] Optionally, the communication interface 2202 can further include a radio frequency transmitter, an antenna, etc. In the case where the communication interface 2202 includes an antenna, the number of antennas can be one or more.

[0414] As a possible design, if the computing device 220 is a standalone device, the communication interface 2202 can include a receiver and a transmitter. The receiver and the transmitter can be the same component, or different components. When the receiver and the transmitter are the same component, the component can be referred to as a transceiver.

[0415] As yet another possible design, if the computing device 220 is a chip or a circuit, the communication interface 2202 can include an input interface and an output interface, which can be the same interface, or can be different interfaces.

[0416] Optionally, the functions of the communication interface 2202 can be implemented by a transceiver circuit or a dedicated chip of transceiver.

[0417] The memory 2203 is configured to provide a storage space in which data such as an operating system and a computer program can be stored. The memory 2203 can be one or a combination of a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), or a compact disc read-only memory (CD-ROM).

[0418] It should be noted that the functions and actions of the modules or units in the computing device 220 listed above are only exemplary.

[0419] The functional units in the computing device 220 can be used to implement the methods implemented by the computing device in the foregoing test method, for example Figure 4 the methods implemented by the computing device in the test method.

[0420] Optionally, the processor 2201 can be a processor specially used for executing the foregoing methods (conveniently distinguished as a special-purpose processor), or can be a processor that executes the foregoing methods by calling computer programs (conveniently distinguished as a special-purpose processor). Optionally, the at least one processor can include both a special-purpose processor and a general-purpose processor.

[0421] Optionally, in the case where the computing device 220 includes at least one memory 2203, if the processor 2201 implements the foregoing test method by calling a computer program, the computer program can be stored in the memory 2203.

[0422] The chip includes a logic circuit and a communication interface. The communication interface is configured to receive a signal or send a signal. The logic circuit is configured to receive a signal or send a signal through the communication interface. The chip is used to implement the foregoing test method, for example Figure 4 the test method.

[0423] The embodiments of the present application further provide a computer readable storage medium, wherein instructions are stored in the computer readable storage medium, and when the instructions are run on at least one processor (or computing device), the foregoing test method is implemented, for example Figure 4 the test method shown.

[0424] The embodiments of the present application further provide a computer program product, which comprises computer instructions for implementing the foregoing test method, for example Figure 4 the test method shown.

[0425] The embodiments of the present application further provide a terminal, which comprises the foregoing test device 210 and / or computing device 220.

[0426] As a possible implementation, the terminal comprises a terminal node, wherein the terminal can be a smart terminal or a transportation tool such as a vehicle, a drone or a robot.

[0427] In the description of the present application, the terms "center", "upper", "lower", "vertical", "horizontal", "inner", "outer", "left", "side" and the like indicate the orientation or positional relationship shown in the drawings, and are only for the convenience of describing the present application and simplifying the description, and do not indicate or imply that the device or element referred to must have a particular orientation, be constructed and operated in a particular orientation, and therefore cannot be understood as a limitation on the present application.

[0428] In the embodiments of the present application, the "end" appearing in the words "one end", "the other end", "left end", "right end", "upper end", "lower end", "connecting end" and the like is not limited to the end, end point or end face, but also includes a part extending an axial distance and / or radial distance from the end, end point or end face on the device or element to which the end, end point or end face belongs.

[0429] In the embodiments of the present application, the words "exemplary" or "for example" are used to mean serving as an example, instance, or illustration. Any embodiment or design described herein as "exemplary" or "for example" should not be construed as being preferred or advantageous over other embodiments or designs. Rather, the use of the words "exemplary" or "for example" is intended to present concepts in a concrete manner.

[0430] The "at least one" mentioned in the embodiments of the present application refers to one or more, and "multiple" refers to two or more. "At least one of the following" or the like refers to any combination of these items, including any combination of single or multiple items. For example, at least one of a, b, or c can represent: a, b, c, (a and b), (a and c), (b and c), or (a and b and c), where a, b, and c can be single or multiple. "And / or" describes the relationship between the associated objects, indicating that there can be three relationships, for example, A and / or B, which can represent three cases: A exists alone, A and B exist together, and B exists alone, where A and B can be singular or plural. The character " / " generally represents an "or" relationship between the associated objects.

[0431] In addition, unless otherwise stated, the ordinal numbers "first", "second", etc. used in the embodiments of the present application are used to distinguish a plurality of objects, and are not used to limit the order, time sequence, priority or importance of the plurality of objects. For example, the first angle measurement data and the second angle measurement data are only used to facilitate the description of fresh parameters in different embodiments, and do not indicate the difference in operation, importance, data content, etc.

[0432] A person of ordinary skill in the art can understand that all or part of the steps of the above-mentioned embodiments can be completed by hardware, or by a program instructing relevant hardware to complete, and the program can be stored in a computer readable storage medium, such as a read-only memory, a magnetic disk or an optical disk.

Claims

1. A testing method, characterized in that, The method includes: A first image is acquired, which is an image from the driver's perspective of the test vehicle. The test vehicle includes a first HUD, which is used to generate a projected image. The projected image is located at the driver's perspective of the test vehicle. The first image includes real-scene information and virtual image information. The real-scene information consists of elements in the actual driving environment of the test vehicle, and the virtual image information consists of elements in the projected image that are related to the real-scene information. Based on the first image, a first feature of the virtual image information and a second feature of the real scene information are obtained. The first feature is used as an attribute representation of the virtual image information, and the second feature is used as an attribute representation of the real scene information. The attribute includes position and / or direction. The virtual image information includes cruise indication information and / or attribute indication information. The attribute indication information includes one or more of the following: pedestrian identification, obstacle identification, preceding vehicle identification, or lane line identification. Based on the first feature and the second feature, the fitting degree evaluation result of the first HUD is determined, and the fitting degree evaluation result is used to indicate the degree of fitting between the projected image of the first HUD and the real driving environment.

2. The method according to claim 1, characterized in that, The determination of the fit assessment result of the first HUD based on the first feature and the second feature includes: Determine the first degree of fit between the first feature and the second feature; Based on the first fit, the fit evaluation result of the first HUD is determined.

3. The method according to claim 2, characterized in that, The first degree of fit is related to the positions of the first feature and the second feature.

4. The method according to claim 2 or 3, characterized in that, The first degree of fit is also related to the type of virtual image information, which includes cruise indication type.

5. The method according to any one of claims 1-4, characterized in that, The virtual image information includes regular icons and / or irregular icons; The first feature of the rule icon is the geometric center point of the rule icon, and the rule icon includes icons that are themselves regular geometric shapes and icons whose outer borders are regular geometric shapes; The first feature of the irregular icon is the key element in the irregular icon, which includes key lines and / or key points.

6. The method according to claim 5, characterized in that, The irregular icons include one or more of curve icons, arrow icons, or road surface coverage icons, and the curve icons include lane line icons; The first feature of the curve identifier is one or more first key points of the curve identifier, wherein the first key point is the connection point between the tangent of the curve identifier and the curve identifier; The first feature of the arrow marker is the second key point or the first key line of the arrow marker. The second key point includes the starting point and the ending point of the arrow marker, and the first key line is the ray from the starting point to the ending point. The first feature of the road surface covering mark is a plurality of third key points of the road surface covering mark, the plurality of third key points including the center point of the boundary on the side closer to the vehicle and the center point of the boundary on the side farther from the vehicle.

7. The method according to any one of claims 1-4, characterized in that, The virtual image information includes cruise indication information. The first feature of the cruise indication information is the fourth key point or the second key line of the cruise indication information. The fourth key point includes the start point and the end point of the cruise indication information. The second key line is a ray from the start point to the end point.

8. The method according to claim 3, characterized in that, Determining the first degree of fit between the first feature and the second feature includes: Based on the first feature, the second feature, and the starting point of the driver's view, a first degree of fit between the first feature and the second feature is determined. The first degree of fit is used to measure the fit deviation between real scene information and virtual image information. The first degree of fit is the value of the angle formed by the first feature and the second feature relative to the starting point of the driver's view. Based on the first fit, a first degree of fit between the first feature and the second feature is determined.

9. The method according to claim 6, characterized in that, The virtual image information is a lane line marker, the first feature is one or more first key points of the lane line marker, and the determination of the first fit between the first feature and the second feature based on the first feature, the second feature, and the starting point of the driver's view includes: Based on the first feature, the second feature, and the starting point of the driver's perspective, a second degree of fit between the first feature and the second feature is determined, and the second degree of fit includes multiple factors. The third fit degree with the largest value among the plurality of second fit degrees is determined as the first fit degree.

10. The method according to claim 8 or 9, characterized in that, The higher the value of the first fit, the lower the degree of fit.

11. The method according to claim 8 or 9, characterized in that, The first fit includes a first horizontal fit and a first vertical fit.

12. The method according to any one of claims 1-11, characterized in that, The second feature of the real-scene information is the determination of the target projection position based on the virtual image information, where the target projection position is the position where the virtual image information is expected to be projected.

13. The method according to claim 4, characterized in that, The first degree of fit is determined based on the target conditions satisfied by the virtual image information.

14. The method according to claim 13, characterized in that, The virtual image information is cruise guidance information, which includes arrow markers, and the target conditions include at least one of the following: The arrow indicates the end point within the target road, which includes the target lane; The end point indicated by the arrow is located within the target lane; The extended line extending in the direction indicated by the arrow intersects the target road, and the distance between the intersection point and the vehicle is less than the first distance; The extension line, which extends in the direction indicated by the arrow, intersects the target lane, and the distance between the intersection point and the vehicle is less than the second distance.

15. The method according to claim 13, characterized in that, The virtual image information is cruise guidance information, which includes road surface coverage markings, and the target conditions include at least one of the following: The center point of the boundary closest to the vehicle is located in front of the vehicle; The center point of the boundary on the side away from the vehicle is located within the target road, which is the estimated road that the vehicle will travel on, and the target road includes the target lane; The center point of the boundary on the side furthest from the vehicle is located within the target lane; The road surface markings have a natural shape; The road surface markings change continuously and smoothly.

16. The method according to any one of claims 13-15, characterized in that, The first degree of fit is determined based on the target conditions satisfied by the virtual image information and the first scene, where the first scene belongs to the scene supported by the first HUD.

17. The method according to claim 16, characterized in that, The virtual image information is cruise guidance information, which includes arrow markers. The first scenario includes a target road curvature greater than 0. The target conditions also include a road curve tendency consistent with the target road. The target road is the estimated road that the vehicle will travel on.

18. The method according to any one of claims 13-17, characterized in that, The first degree of fit is determined based on the number of target conditions satisfied by the virtual image information. The more target conditions satisfied by the virtual image information, the higher the degree of fit.

19. The method according to any one of claims 1-18, characterized in that, The method further includes: A test scenario database is constructed, which includes multiple test scenarios. Any two test scenarios differ on at least one of multiple test factors, including road static factors, dynamic traffic participant factors, and environmental factors. The acquisition of the first image includes: The first image is obtained based on the first scene from multiple test scenes in the test scene database.

20. The method according to claim 19, characterized in that, The multiple test factors also include test vehicle-specific factors, which include one or more of the following: vehicle basic information, location information, or motion state information, and the motion state information includes vehicle speed.

21. The method according to claim 19 or 20, characterized in that, The road static factors include one or more of the following: intersection shape information, road geometry information, road surface condition information, landmark information, or lane line condition information; The dynamic traffic participant factors include one or more of the relative positional relationships between motor vehicles, non-motorized vehicles, pedestrians and the test vehicle. The environmental factors include one or more of time, brightness, or weather.

22. The method according to claim 2, characterized in that, The first image includes a second image and a third image, wherein the second image and the third image are images of the first image at a time interval of a first time. The first degree of fit includes a second degree of fit and a third degree of fit. The second degree of fit is the degree of fit between the virtual image information and the real scene information in the second image, and the third degree of fit is the degree of fit between the virtual image information and the real scene information in the third image. The determination of the fit assessment result of the first HUD based on the first fit degree includes: Based on the second fit degree and the third fit degree, the fit degree evaluation result of the first HUD is determined, and the fit degree evaluation result includes the jitter evaluation result.

23. The method according to claim 22, characterized in that, The jitter evaluation results include jitter amplitude and jitter frequency. The jitter amplitude is used to indicate the maximum fluctuation of the fit of the first image. The jitter amplitude includes single-cycle jitter amplitude and jitter distribution dispersion. The single-cycle jitter amplitude includes a first jitter amplitude obtained based on the second fit degree and the third fit degree, wherein the second fit degree and the third fit degree are the maximum and minimum values ​​within the first time period, respectively.

24. A testing system, characterized in that, The testing system includes a computing device, an image acquisition device, and a test vehicle, wherein the computing device is used to implement the testing method according to any one of claims 1-23; The image acquisition device is installed on the line corresponding to the driver's lower field of view (LDA) in the driver's seat of the test vehicle. The lens of the image acquisition device faces the outside of the test vehicle. The image acquisition device is used to acquire video or images from the driver's perspective of the test vehicle. The test vehicle includes a first HUD for generating a projected image located at the driver's viewpoint of the test vehicle.

25. The testing system according to claim 24, characterized in that, The computing device is connected to the test vehicle and is used to acquire relevant data of the first HUD of the test vehicle under test conditions.

26. A testing apparatus, characterized in that, The device includes: A communication unit is configured to acquire a first image, which is an image from the driver's perspective of the test vehicle. The test vehicle includes a first HUD, which is used to generate a projected image. The projected image is located at the driver's perspective of the test vehicle. The first image includes real-scene information and virtual image information. The real-scene information consists of elements in the actual driving environment of the test vehicle, and the virtual image information consists of elements in the projected image that are related to the real-scene information. The processing unit is configured to obtain, based on the first image, a first feature of the virtual image information and a second feature of the real scene information, wherein the first feature is used as an attribute representation of the virtual image information and the second feature is used as an attribute representation of the real scene information, and the attribute includes position and / or direction; the virtual image information includes cruise indication information and / or attribute indication information, and the attribute indication information includes one or more of pedestrian markings, obstacle markings, preceding vehicle markings, or lane line markings; Based on the first feature and the second feature, the fitting degree evaluation result of the first HUD is determined, and the fitting degree evaluation result is used to indicate the degree of fitting between the projected image of the first HUD and the real driving environment.

27. A chip, characterized in that, The chip includes a processor and a communication interface; The communication interface is used to receive and / or send data, and / or the communication interface is used to provide input and / or output to the processor; The processor is used to implement the method according to any one of claims 1-23.

28. A computing device, characterized in that, The computing device includes a memory and a processor, the memory storing computer instructions, and the processor being used to invoke the computer instructions to implement the method of any one of claims 1-23.

29. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores instructions that, when executed on at least one processor, implement the method as described in any one of claims 1-23.

Images