Camera simulation test method, device, equipment and readable storage medium

Abstract

The present invention provides a camera simulation test method, device, equipment and readable storage medium. The camera simulation test method includes: setting the initial posture of a virtual camera according to the installation posture range of a camera to be tested; obtaining simulation data of the virtual camera identifying a target in a simulation scene; obtaining test data of the camera to be tested identifying a target in a simulation scene; comparing the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current posture; adjusting the posture of the virtual camera according to a preset step size and a preset angle; and outputting a summary report of the recognition accuracy of the camera to be tested in each posture. Through the present invention, the recognition accuracy of each posture of the camera within the working posture range can be comprehensively and objectively tested and evaluated, the installation posture with the best recognition accuracy of the camera can be obtained, and reference suggestions for the best installation posture of the camera can be provided.

Description

Technical Field

[0001] The present invention relates to the field of simulation testing, and in particular to a camera simulation testing method, device, equipment and readable storage medium. Background Art

[0002] In the simulation test of the vehicle-mounted smart camera, the two most commonly used solutions are video dark box and video injection. Both solutions are based on the X-coordinate, Y-coordinate, Z-coordinate, pitch angle, yaw angle and roll angle provided by the provider of the camera to be tested, and the installation position and angle parameters of the virtual camera are configured in the simulation test scene accordingly, and then the simulation scene is started for testing. However, the deviation of the camera installation posture will directly affect the vehicle's perception of the external environment, and then affect the judgment and decision-making of the vehicle's automatic driving system. Using a single installation posture parameter to simulate the camera cannot form a comprehensive and objective evaluation of the camera, thereby weakening the advantages of simulation testing. Summary of the invention

[0003] The main purpose of the present invention is to provide a camera simulation test method, device, equipment and readable storage medium, aiming to solve the technical problems that the simulation test of the camera using a single installation posture is not comprehensive and objective, and the optimal installation posture parameters of the camera cannot be determined.

[0004] In a first aspect, the present invention provides a camera simulation test method, the camera simulation test method comprising:

[0005] In the simulation test environment, the initial position of the virtual camera is set according to the installation position range of the camera to be tested;

[0006] Acquire simulation data of the virtual camera identifying the target object in the simulation scene;

[0007] Acquire test data of the camera to be tested identifying the target object in the simulation scene;

[0008] Comparing the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position;

[0009] Adjust the posture of the virtual camera according to the preset step length and the preset angle. If the installation posture range of the camera to be tested has not been tested, return to the step of obtaining simulation data of the virtual camera identifying the target object in the simulation scene;

[0010] The recognition accuracy of the camera to be tested at each position is summarized, and a summary report of the recognition accuracy of the camera to be tested at each position is output.

[0011] Optionally, before obtaining the simulation data of the virtual camera identifying the target object in the simulation scene, the method includes:

[0012] A simulation scene is selected from a simulation scene library according to the recognition characteristics of the camera to be tested, wherein the recognition characteristics include weather conditions, lighting conditions and target object type.

[0013] Optionally, before obtaining the test data of the camera to be tested identifying the target object in the simulation scene, the method includes:

[0014] Based on the video dark box of the simulation test bench, set the installation position of the camera to be tested.

[0015] Optionally, comparing the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position and posture includes:

[0016] Comparing the simulation data with the test data, obtaining the recognition accuracy of each target object in the simulation scene at the current position of the camera to be tested;

[0017] The recognition accuracy rates of each target object belonging to the same target object type are averaged, and the recognition accuracy rates of each target object type are calculated respectively;

[0018] According to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy of different target types, and the recognition accuracy of the camera to be tested in the current position is calculated.

[0019] Optionally, the posture includes an X coordinate, a Y coordinate, a Z coordinate, a yaw angle, a roll angle, and a pitch angle, and adjusting the posture of the virtual camera according to a preset step size and a preset angle includes:

[0020] According to the preset step length, adjust the X coordinate value, Y coordinate value and Z coordinate value of the virtual camera respectively;

[0021] According to the preset angles, the yaw angle, roll angle and pitch angle of the virtual camera are adjusted respectively.

[0022] Optionally, after summarizing the recognition accuracy of the camera to be tested at each position and outputting a summary report of the recognition accuracy of the camera to be tested at each position and posture, the method further includes:

[0023] In the order of recognition accuracy from high to low, a preset number of postures are selected from the data after summarizing the recognition accuracy of the camera to be tested at each posture as the posture data set for a new round of testing of the camera to be tested;

[0024] If the pose data set has not been tested, selecting a pose from the pose data set as a pose to be tested;

[0025] Based on the posture to be tested, the posture of the virtual camera is adjusted according to the new preset step size and the new preset angle. If the preset range of the posture to be tested has not been tested, the step of obtaining simulation data for the virtual camera to identify the target object in the simulation scene is returned to execute, and the new preset step size is smaller than the preset step size, and the new preset angle is smaller than the preset angle.

[0026] In a second aspect, the present invention further provides a camera simulation test device, the camera simulation test device comprising:

[0027] A setting module is used to set the initial position of the virtual camera according to the installation position range of the camera to be tested in the simulation test environment;

[0028] A first acquisition module is used to acquire simulation data of a virtual camera identifying a target object in a simulation scene;

[0029] The second acquisition module is used to obtain test data of the camera to be tested identifying the target object in the simulation scene;

[0030] A comparison module, used to compare the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position;

[0031] An adjustment module, used for adjusting the posture of the virtual camera according to a preset step length and a preset angle, and if the installation posture range of the camera to be tested has not been tested, returning to the step of obtaining simulation data of the virtual camera identifying the target object in the simulation scene;

[0032] The output module is used to summarize the recognition accuracy of the camera to be tested at each position and output a summary report of the recognition accuracy of the camera to be tested at each position and position.

[0033] Optionally, the comparison module is used to:

[0034] Comparing the simulation data with the test data, obtaining the recognition accuracy of each target object in the simulation scene at the current position of the camera to be tested;

[0035] The recognition accuracy rates of each target object belonging to the same target object type are averaged, and the recognition accuracy rates of each target object type are calculated respectively;

[0036] According to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy of different target types, and the recognition accuracy of the camera to be tested in the current position is calculated.

[0037] In a third aspect, the present invention also provides a camera simulation test device, comprising a processor, a memory, and a camera simulation test program stored in the memory and executable by the processor, wherein when the camera simulation test program is executed by the processor, the steps of the camera simulation test method described above are implemented.

[0038] In a fourth aspect, the present invention further provides a readable storage medium, on which a camera simulation test program is stored, wherein when the camera simulation test program is executed by a processor, the steps of the camera simulation test method as described above are implemented.

[0039] In the present invention, in a simulation test environment, the initial posture of a virtual camera is set according to the installation posture range of a camera to be tested; simulation data of the virtual camera identifying a target object in a simulation scene is obtained; test data of the camera to be tested identifying a target object in a simulation scene is obtained; the simulation data and the test data are compared to obtain the recognition accuracy of the camera to be tested in the current posture; the posture of the virtual camera is adjusted according to a preset step size and a preset angle, and if the installation posture range of the camera to be tested has not been tested, the step of obtaining the simulation data of the virtual camera identifying the target object in the simulation scene is returned to execute; the recognition accuracy of the camera to be tested in each posture is summarized, and a summary report of the recognition accuracy of the camera to be tested in each posture is output. The present invention sets the initial posture of the virtual camera according to the installation posture range of the camera to be tested in a simulation test environment. The installation posture range of the camera is usually provided by the camera manufacturer. The simulation data of the virtual camera identifying the target object in the simulation scene and the test data of the camera to be tested identifying the target object in the simulation scene are respectively obtained, which are usually called "true value" and "test value" respectively. By comparing the "true value" and the "test value", the recognition accuracy of the target object in the simulation scene in the current posture of the camera to be tested can be obtained, and then according to the preset step size and preset angle, Adjust the posture of the virtual camera, and perform traversal testing on the camera to be tested within the installation posture range, so as to obtain the recognition accuracy of the camera to be tested in each posture, and then summarize and output a simulation test summary report of the recognition accuracy of the camera to be tested in each posture. Through the present invention, the recognition accuracy of each posture of the camera within the working posture range can be efficiently tested and evaluated in a comprehensive and objective manner, and the installation posture of the camera with the best recognition accuracy can be obtained, thereby providing camera manufacturers and related parties with reference suggestions for the best installation posture of the camera, further improving the recognition accuracy of the camera. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1A schematic diagram of the hardware structure of an embodiment of a camera simulation test device of the present invention;

[0041] Figure 2 A schematic diagram of a flow chart of an embodiment of a camera simulation test method of the present invention;

[0042] Figure 3 for Figure 2 A detailed flow chart of step S40;

[0043] Figure 4 It is a schematic diagram of the process of a new round of testing in an embodiment of the camera simulation testing method of the present invention;

[0044] Figure 5 Schematic diagram of functional modules of an embodiment of a camera simulation test device of the present invention.

[0045] The realization of the purpose, functional features and advantages of the present invention will be further explained in conjunction with embodiments and with reference to the accompanying drawings. DETAILED DESCRIPTION

[0046] It should be understood that the specific embodiments described herein are only used to explain the present invention, and are not used to limit the present invention.

[0047] In a first aspect, an embodiment of the present invention provides a camera simulation test device.

[0048] Reference Figure 1 , Figure 1 The hardware structure diagram of an embodiment of the camera simulation test device of the present invention. In the embodiment of the present invention, the camera simulation test device may include a processor 1001 (e.g., a central processing unit, CPU), a communication bus 1002, a user interface 1003, a network interface 1004, and a memory 1005. Among them, the communication bus 1002 is used to realize the connection and communication between these components; the user interface 1003 may include a display screen (Display), an input unit such as a keyboard (Keyboard); the network interface 1004 may optionally include a standard wired interface, a wireless interface (such as a wireless fidelity WIreless-FIdelity, WI-FI interface); the memory 1005 may be a high-speed random access memory (random access memory, RAM), or a stable memory (non-volatile memory), such as a disk storage, and the memory 1005 may optionally be a storage device independent of the aforementioned processor 1001. Those skilled in the art will understand that Figure 1 The hardware structure shown in the figure does not constitute a limitation of the present invention, and may include more or less components than those shown in the figure, or combine certain components, or arrange the components differently.

[0049] Continue to refer to Figure 1 , Figure 1 The memory 1005 as a computer storage medium may include an operating system, a network communication module, a user interface module, and a camera simulation test program. The processor 1001 may call the camera simulation test program stored in the memory 1005 and execute the camera simulation test method provided in the embodiment of the present invention.

[0050] In a second aspect, an embodiment of the present invention provides a camera simulation test method.

[0051] In order to more clearly demonstrate the camera simulation test method provided in the embodiment of the present application, the application scenario of the camera simulation test method provided in the embodiment of the present application is first introduced.

[0052] The camera simulation test method provided in the embodiment of the present application is used to obtain test data such as the recognition accuracy of the camera for different types of targets in various installation positions through simulation testing, so as to perform performance testing on the camera.

[0053] In one embodiment, referring to Figure 2 , Figure 2 FIG. 1 is a flow chart of an embodiment of a camera simulation test method of the present invention. Figure 2 As shown, the camera simulation test method includes:

[0054] Step S10: in a simulation test environment, setting the initial posture of the virtual camera according to the installation posture range of the camera to be tested.

[0055] In this embodiment, the simulation test environment is usually composed of a simulation test bench, which includes a graphics workstation, a host computer, a slave computer, etc. equipped with virtual simulation software. The installation position range of the camera to be tested is usually provided by the camera manufacturer, which specifies the installation position range suitable for the camera to work. Within the installation position range of the camera to be tested, the initial position of the virtual camera is set, and the performance test of the camera to be tested is implemented in the simulation environment.

[0056] Step S20, obtaining simulation data of the virtual camera identifying the target object in the simulation scene.

[0057] In this embodiment, simulation data of a virtual camera identifying a target object in a simulation scene is obtained through simulation test software. The simulation data is usually referred to as a “true value”.

[0058] Step S30, obtaining test data of the camera to be tested identifying the target object in the simulation scene.

[0059] In this embodiment, the test data of the camera to be tested recognizing the target object in the simulated scene is obtained through simulation test software. The test data is usually referred to as "test value". Based on the camera simulation test scheme of video darkroom and video injection, the synchronization of the image recognition of the camera to be tested and the virtual camera can be achieved, that is, the camera to be tested and the virtual camera can recognize the same target object scene image in real time.

[0060] Step S40, comparing the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position.

[0061] In this embodiment, the simulation data and the test data are compared, that is, the "true value" and the "test value" are compared to obtain the recognition accuracy of the camera to be tested in the current position. According to the recognition characteristics of the camera and the specific needs of the simulation test, the simulation data and the test data may specifically include the distance of the target object relative to the camera, the width of the target object, and the type of the target object.

[0062] Step S50, adjusting the posture of the virtual camera according to the preset step size and the preset angle. If the installation posture range of the camera to be tested has not been tested, returning to the step of obtaining simulation data for the virtual camera to identify the target object in the simulation scene.

[0063] In this embodiment, the preset step size and the preset angle are set according to specific test needs. Within the installation posture range of the camera to be tested, the preset step size and the preset angle are used to gradually adjust the posture of the virtual camera to perform a traversal test on the camera to be tested.

[0064] Step S60, summarizing the recognition accuracy of the camera to be tested at each position and posture, and outputting a summary report of the recognition accuracy of the camera to be tested at each position and posture.

[0065] In this embodiment, after summarizing the recognition accuracy of the camera to be tested in various positions, the recognition accuracy of the camera to be tested in various positions can be known, and a summary report of the recognition accuracy of the camera to be tested in various positions can be output. This can provide camera manufacturers and related parties with reference suggestions for the best installation position of the camera, and efficiently conduct comprehensive and objective test and evaluation of the recognition accuracy of the camera in various positions within the working position range.

[0066] In the present embodiment, within the installation posture range of the camera to be tested, the posture of the virtual camera is adjusted according to a preset step size and a preset angle, and a traversal simulation test is performed on the camera to be tested. By respectively obtaining simulation data of the virtual camera identifying the target object in the simulation scene and test data of the camera to be tested identifying the target object in the simulation scene, that is, the "true value" and the "test value", the "true value" and the "test value" are compared to obtain the recognition accuracy of the camera to be tested for the target object in the simulation scene in the current posture, and further obtain the recognition accuracy of the camera to be tested in each posture, thereby achieving a comprehensive and objective test and evaluation of the recognition accuracy of each posture of the camera within the working posture range, and obtaining the installation posture with the best recognition accuracy of the camera, thereby providing camera manufacturers and related parties with reference suggestions for the best installation posture of the camera, and further improving the recognition accuracy of the camera.

[0067] Furthermore, in one embodiment, before step S20, the following steps are included:

[0068] A simulation scene is selected from a simulation scene library according to the recognition characteristics of the camera to be tested, wherein the recognition characteristics include weather conditions, lighting conditions and target object type.

[0069] In this embodiment, different cameras are good at identifying different weather conditions, lighting conditions and target types. For example, some cameras are good at identifying lane lines and vehicles, and work better under good weather conditions. Therefore, according to the recognition of different weather conditions, lighting conditions and target types by the camera to be tested, the corresponding simulation scene is selected to achieve better targeting and simulated testing of the camera to be tested.

[0070] Furthermore, in one embodiment, before step S30, the following steps are included:

[0071] Based on the video dark box of the simulation test bench, set the installation position of the camera to be tested.

[0072] In this embodiment, the synchronization of the simulated scene recognized by the camera to be tested and the virtual camera is achieved through a video darkbox. The principle is to set a display screen in the video darkbox of the simulation test bench, and install the camera to be tested in front of the display screen so that the recognition field of view of the camera to be tested coincides with the display screen, and the optical axis of the camera to be tested is perpendicular to the display screen. The image of the simulated scene recognized by the virtual camera is synchronized to the display screen in the video darkbox in real time through the simulation software, that is, each time after adjusting the posture of the virtual camera, the image recognized by the virtual camera will be synchronized to the display screen in the video darkbox in real time, and the camera to be tested will recognize the display screen, thereby achieving synchronization of the simulated scene recognized by the camera to be tested and the virtual camera.

[0073] Further, in one embodiment, referring to Figure 3 , Figure 3 for Figure 2 The detailed flow chart of step S40 is as follows: Figure 3 As shown, step S40 includes:

[0074] Step S401, comparing the simulation data with the test data to obtain the recognition accuracy of each target object in the simulation scene at the current position of the camera to be tested;

[0075] Step S402, taking an average of the recognition accuracy rates of each target object belonging to the same target object type, and calculating the recognition accuracy rate of each target object type respectively;

[0076] Step S403, according to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy rates of different target types, and the recognition accuracy rate of the camera to be tested in the current position is calculated.

[0077] In this embodiment, there are usually multiple types of multiple targets in a simulation scene. After comparing the simulation data and the test data to obtain the recognition accuracy of the multiple types of multiple targets in the simulation scene, the accuracy of the targets of the same type is averaged to obtain the recognition accuracy of each type. Then, according to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy of different target types. After calculation, the comprehensive recognition accuracy of the camera to be tested for various types of targets in the current position is obtained. For example, a camera to be tested is mostly used to identify lane lines. For the target types of the camera to be tested: lane lines, vehicles, and pedestrians, corresponding weights of 50%, 30%, and 20% can be assigned respectively. The specific weight setting can be performed according to the recognition characteristics of the camera and the test requirements.

[0078] Further, in one embodiment, the posture includes an X coordinate, a Y coordinate, a Z coordinate, a yaw angle, a roll angle, and a pitch angle, and adjusting the posture of the virtual camera according to a preset step size and a preset angle includes:

[0079] According to the preset step length, adjust the X coordinate value, Y coordinate value and Z coordinate value of the virtual camera respectively;

[0080] According to the preset angles, the yaw angle, roll angle and pitch angle of the virtual camera are adjusted respectively.

[0081] In this embodiment, the posture of the camera includes X-coordinate, Y-coordinate, Z-coordinate, yaw angle, roll angle and pitch angle, that is, six degrees of freedom. When adjusting the posture of the virtual camera, the X-coordinate value, Y-coordinate value and Z-coordinate value of the virtual camera are adjusted respectively according to a preset step size, and the yaw angle, roll angle and pitch angle of the virtual camera are adjusted respectively according to a preset angle, so as to test multiple postures of the camera to be tested within the installation posture range.

[0082] Further, in one embodiment, referring to Figure 4 , Figure 4 FIG. 1 is a flow chart of a new round of testing in an embodiment of a camera simulation testing method of the present invention. Figure 4 As shown, after step S60, a step S70 of performing a new round of testing on the camera to be tested is further included, and step S70 includes:

[0083] Step S701, selecting a preset number of positions from the data after summarizing the recognition accuracy of the camera to be tested at each position in the order of recognition accuracy from high to low, as a position data set for a new round of testing of the camera to be tested;

[0084] Step S702, if the pose data set has not been tested, selecting a pose from the pose data set as a pose to be tested;

[0085] Step S703, based on the posture to be tested, adjust the posture of the virtual camera according to the new preset step size and the new preset angle. If the preset range of the posture to be tested has not been tested, return to the step of obtaining simulation data of the virtual camera identifying the target object in the simulation scene, and the new preset step size is smaller than the preset step size, and the new preset angle is smaller than the preset angle.

[0086] In this embodiment, considering the efficiency of the traversal test, the preset step size and the preset angle in step S50 may be set too large. Therefore, further tests can be performed to obtain a more accurate posture, which is equivalent to a new round of tests. In the data after the recognition accuracy of each posture is summarized, a certain number of postures are selected in order of recognition accuracy from high to low. In the vicinity of the selected posture to be tested, further traversal tests are performed according to the new preset step size and the new preset angle to obtain the posture of the camera with better recognition accuracy.

[0087] In a third aspect, an embodiment of the present invention further provides a camera simulation test device.

[0088] Reference Figure 5 , Figure 5 Schematic diagram of functional modules of an embodiment of a camera simulation test device of the present invention.

[0089] In this embodiment, the camera simulation test device includes:

[0090] A setting module 10 is used to set the initial position of the virtual camera according to the installation position range of the camera to be tested in the simulation test environment;

[0091] A first acquisition module 20 is used to acquire simulation data of a virtual camera identifying a target object in a simulation scene;

[0092] The second acquisition module 30 is used to obtain test data of the camera to be tested identifying the target object in the simulation scene;

[0093] A comparison module 40 is used to compare the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position;

[0094] The adjustment module 50 is used to adjust the posture of the virtual camera according to a preset step length and a preset angle. If the installation posture range of the camera to be tested has not been tested, the step of returning to execute the step of obtaining simulation data of the virtual camera identifying the target object in the simulation scene;

[0095] The output module 60 is used to summarize the recognition accuracy of the camera to be tested at each position and output a summary report of the recognition accuracy of the camera to be tested at each position and position.

[0096] Furthermore, in one embodiment, the camera simulation test device further includes a selection module, which is used to:

[0097] A simulation scene is selected from a simulation scene library according to the recognition characteristics of the camera to be tested, wherein the recognition characteristics include weather conditions, lighting conditions and target object type.

[0098] Furthermore, in one embodiment, the camera simulation test device further includes a posture setting module, which is used to:

[0099] Based on the video dark box of the simulation test bench, set the installation position of the camera to be tested.

[0100] Furthermore, in one embodiment, the comparison module 40 is used to:

[0101] Comparing the simulation data with the test data, obtaining the recognition accuracy of each target object in the simulation scene at the current position of the camera to be tested;

[0102] The recognition accuracy rates of each target object belonging to the same target object type are averaged, and the recognition accuracy rates of each target object type are calculated respectively;

[0103] According to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy of different target types, and the recognition accuracy of the camera to be tested in the current position is calculated.

[0104] Furthermore, in one embodiment, the adjustment module 50 is used to:

[0105] According to the preset step length, adjust the X coordinate value, Y coordinate value and Z coordinate value of the virtual camera respectively;

[0106] According to the preset angles, the yaw angle, roll angle and pitch angle of the virtual camera are adjusted respectively.

[0107] Furthermore, in one embodiment, the camera simulation test device further includes a posture adjustment module, which is used to:

[0108] In the order of recognition accuracy from high to low, a preset number of postures are selected from the data after summarizing the recognition accuracy of the camera to be tested at each posture as the posture data set for a new round of testing of the camera to be tested;

[0109] If the pose data set has not been tested, selecting a pose from the pose data set as a pose to be tested;

[0110] Based on the posture to be tested, the posture of the virtual camera is adjusted according to the new preset step size and the new preset angle. If the preset range of the posture to be tested has not been tested, the step of obtaining simulation data for the virtual camera to identify the target object in the simulation scene is returned to execute, and the new preset step size is smaller than the preset step size, and the new preset angle is smaller than the preset angle.

[0111] Among them, the functional implementation of each module in the above-mentioned camera simulation test device corresponds to the various steps in the above-mentioned camera simulation test method embodiment, and its functions and implementation processes will not be repeated here one by one.

[0112] In a fourth aspect, an embodiment of the present invention further provides a readable storage medium.

[0113] A camera simulation test program is stored on the readable storage medium of the present invention, wherein when the camera simulation test program is executed by a processor, the steps of the camera simulation test method as described above are implemented.

[0114] Among them, the method implemented when the camera simulation test program is executed can refer to the various embodiments of the camera simulation test method of the present invention, and will not be repeated here.

[0115] It should be noted that, in this article, the terms "include", "comprises" or any other variations thereof are intended to cover non-exclusive inclusion, so that a process, method, article or system including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or system. In the absence of further restrictions, an element defined by the sentence "comprises a ..." does not exclude the existence of other identical elements in the process, method, article or system including the element.

[0116] The serial numbers of the above embodiments of the present invention are only for description and do not represent the advantages or disadvantages of the embodiments.

[0117] Through the description of the above implementation methods, those skilled in the art can clearly understand that the above-mentioned embodiment methods can be implemented by means of software plus a necessary general hardware platform, and of course by hardware, but in many cases the former is a better implementation method. Based on such an understanding, the technical solution of the present invention is essentially or the part that contributes to the prior art can be embodied in the form of a software product, which is stored in a storage medium (such as ROM / RAM, magnetic disk, optical disk) as described above, and includes a number of instructions for a terminal device to execute the methods described in each embodiment of the present invention.

[0118] The above are only preferred embodiments of the present invention, and are not intended to limit the patent scope of the present invention. Any equivalent structure or equivalent process transformation made using the contents of the present invention specification and drawings, or directly or indirectly applied in other related technical fields, are also included in the patent protection scope of the present invention.

Claims

1. A camera simulation test method, characterized in that: The camera simulation test method comprises: In the simulation test environment, the initial position of the virtual camera is set according to the installation position range of the camera to be tested; Acquire simulation data of the virtual camera identifying the target object in the simulation scene; Acquire test data of the camera to be tested identifying the target object in the simulation scene; Comparing the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position; Adjust the posture of the virtual camera according to the preset step length and the preset angle. If the installation posture range of the camera to be tested has not been tested, return to the step of obtaining simulation data of the virtual camera identifying the target object in the simulation scene; Summarize the recognition accuracy of the camera to be tested at each position and output a summary report of the recognition accuracy of the camera to be tested at each position and Before obtaining the simulation data of the virtual camera identifying the target object in the simulation scene, the method includes: A simulation scene is selected from a simulation scene library according to the recognition characteristics of the camera to be tested, wherein the recognition characteristics include weather conditions, lighting conditions and target object type.

2. The camera simulation test method according to claim 1, characterized in that: Before obtaining the test data of the camera to be tested identifying the target object in the simulation scene, the method includes: Based on the video dark box of the simulation test bench, set the installation position of the camera to be tested.

3. The camera simulation test method according to claim 1, characterized in that: The comparison of the simulation data with the test data to obtain the recognition accuracy of the camera to be tested at the current position includes: Comparing the simulation data with the test data, obtaining the recognition accuracy of each target object in the simulation scene at the current position of the camera to be tested; The recognition accuracy rates of each target object belonging to the same target object type are averaged, and the recognition accuracy rates of each target object type are calculated respectively; According to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy of different target types, and the recognition accuracy of the camera to be tested in the current position is calculated.

4. The camera simulation test method according to claim 1, characterized in that: The posture includes an X coordinate, a Y coordinate, a Z coordinate, a yaw angle, a roll angle, and a pitch angle. Adjusting the posture of the virtual camera according to a preset step size and a preset angle includes: According to the preset step length, adjust the X coordinate value, Y coordinate value and Z coordinate value of the virtual camera respectively; According to the preset angles, the yaw angle, roll angle and pitch angle of the virtual camera are adjusted respectively.

5. The camera simulation test method according to claim 1, characterized in that: After summarizing the recognition accuracy of the camera to be tested at each position and outputting a summary report of the recognition accuracy of the camera to be tested at each position and posture, the following steps are included: In the order of recognition accuracy from high to low, a preset number of postures are selected from the data after summarizing the recognition accuracy of the camera to be tested at each posture as the posture data set for a new round of testing of the camera to be tested; If the pose data set has not been tested, selecting a pose from the pose data set as the pose to be tested; Based on the posture to be tested, the posture of the virtual camera is adjusted according to the new preset step size and the new preset angle. If the preset range of the posture to be tested has not been tested, the step of obtaining simulation data for the virtual camera to identify the target object in the simulation scene is returned to execute, and the new preset step size is smaller than the preset step size, and the new preset angle is smaller than the preset angle.

6. A camera simulation test device, characterized in that: The camera simulation test device comprises: A setting module is used to set the initial position of the virtual camera according to the installation position range of the camera to be tested in the simulation test environment; A first acquisition module is used to acquire simulation data of a virtual camera identifying a target object in a simulation scene; The second acquisition module is used to obtain test data of the camera to be tested identifying the target object in the simulation scene; A comparison module, used to compare the simulation data with the test data to obtain the recognition accuracy of the camera to be tested in the current position; An adjustment module, used for adjusting the posture of the virtual camera according to a preset step length and a preset angle, and if the installation posture range of the camera to be tested has not been tested, returning to the step of obtaining simulation data of the virtual camera identifying the target object in the simulation scene; An output module is used to summarize the recognition accuracy of the camera to be tested at each position and output a summary report of the recognition accuracy of the camera to be tested at each position and position; The camera simulation test device also includes a selection module, which is used to: A simulation scene is selected from a simulation scene library according to the recognition characteristics of the camera to be tested, wherein the recognition characteristics include weather conditions, lighting conditions and target object type.

7. The camera simulation test device according to claim 6, characterized in that: The comparison module is used for: Comparing the simulation data with the test data, obtaining the recognition accuracy of each target object in the simulation scene at the current position of the camera to be tested; The recognition accuracy rates of each target object belonging to the same target object type are averaged, and the recognition accuracy rates of each target object type are calculated respectively; According to the recognition characteristics of the camera to be tested, different weights are assigned to the recognition accuracy of different target types, and the recognition accuracy of the camera to be tested in the current position is calculated.

8. A camera simulation test device, characterized in that: The camera simulation test device includes a processor, a memory, and a camera simulation test program stored in the memory and executable by the processor, wherein when the camera simulation test program is executed by the processor, the steps of the camera simulation test method as described in any one of claims 1 to 5 are implemented.

9. A readable storage medium, characterized in that: The readable storage medium stores a camera simulation test program, wherein when the camera simulation test program is executed by a processor, the steps of the camera simulation test method as described in any one of claims 1 to 5 are implemented.

Images