An intelligent camera startup time testing method and system

Abstract

A method for testing the startup time of an intelligent camera, which relates to the field of testing components of an intelligent networked system. By setting a target object, the reference longitudinal distance from the camera to the target object and the error threshold are obtained; the camera is powered on, and the first timestamp T1 is recorded; then the actual longitudinal distance from the camera to the target object is identified, the variance is calculated by subtracting it from the reference longitudinal distance, and the second timestamp T2 is recorded; when the variance is less than the error threshold, the startup time T = T2 - T1. This application makes the startup time of the camera more accurate; and the whole process can be carried out in real time and automatically without human participation, avoiding human errors and making the detection efficiency and detection accuracy of the startup time higher.

Description

Technical Field

[0001] This application relates to the field of intelligent connected system component testing, and specifically relates to a method and system for testing the startup time of an intelligent camera. Background Art

[0002] The startup time of an intelligent camera refers to the time interval from power-on to stable operation. This time must be short to ensure that the vehicle can quickly enter the intelligent driving mode after the user starts the vehicle. At the same time, this time value is an important basis for intelligent camera fault diagnosis.

[0003] Currently, there are not many testing methods for the startup time of intelligent cameras. The startup time of intelligent cameras is usually estimated based on the values declared by the manufacturer or by roughly using a stopwatch, with low accuracy. Therefore, the startup time cannot be obtained accurately, and there are large human errors. Summary of the Invention

[0004] Embodiments of this application provide a method and system for testing the startup time of an intelligent camera to solve the problem of low accuracy in obtaining the startup time of an intelligent camera in related technologies.

[0005] In a first aspect, a method for testing the startup time of an intelligent camera is provided, including:

[0006] Set a target object, and obtain the reference longitudinal distance from the camera to the target object and the error threshold;

[0007] Power on the camera, and record the first timestamp at the same time;

[0008] Identify the actual longitudinal distance from the camera to the target object, calculate the variance after subtracting it from the reference longitudinal distance, and record the second timestamp at the same time;

[0009] When the variance is less than or equal to the error threshold, subtract the first timestamp from the second timestamp to obtain the startup time.

[0010] In some embodiments, when the variance is greater than the error threshold, the second timestamp is invalid, and an error reporting process is performed.

[0011] In some embodiments, the setting of the target object includes: setting a longitudinal distance range between the target object and the camera, placing the target object within the longitudinal distance range, and the target object facing the camera directly.

[0012] In some embodiments, the longitudinal distance range between the set target and the camera includes: setting the target, and obtaining the maximum value and the minimum value of the longitudinal distance on the principle that the longitudinal distance from the camera to the target is not a null value; the value range between the two is the longitudinal distance range.

[0013] In some embodiments, obtaining the reference longitudinal distance from the camera to the target includes: setting the number of tests, powering on the camera, performing hardware-in-the-loop testing on the camera, obtaining the longitudinal distance from the camera to the target and recording it;

[0014] Power off the camera and then power it on again, repeat the test and record the longitudinal distance each time until the number of tests is reached and the test ends;

[0015] Calculate the mean value of all recorded longitudinal distances as the reference longitudinal distance.

[0016] In some embodiments, the repeated test is realized by building an automated test sequence to achieve fully automatic repeated power-on and power-off tests; after each power-off, after a preset time interval, the camera starts the next power-on.

[0017] In some embodiments, during the process of obtaining the reference longitudinal distance from the camera to the target, if the position, type or attitude of the target is changed, and the longitudinal distance from the camera to the target directly in front of the camera is not a null value, there is no need to obtain it again;

[0018] Otherwise, if the longitudinal distance is a null value and the position, type, attitude or angle of the camera is changed, it needs to be obtained again.

[0019] In some embodiments, obtaining the error threshold includes: subtracting the reference longitudinal distance from all recorded longitudinal distances in sequence, then squaring and summing them to obtain the variance, and taking this variance as the error threshold.

[0020] In some embodiments, the longitudinal distance refers to the distance from the target directly in front of the optical axis of the camera to the camera in a coordinate system with the camera as the origin.

[0021] In a second aspect, an intelligent camera startup time test system is provided, including:

[0022] An acquisition module, configured to set a target, and acquire the reference longitudinal distance and error threshold of the camera;

[0023] A timestamp recording module, configured to record a first timestamp simultaneously when the camera is powered on; and is further configured to identify the actual longitudinal distance from the camera to the target and record a second timestamp;

[0024] A calculation module, configured to calculate the variance after subtracting the actual longitudinal distance from the reference longitudinal distance; when the variance is less than or equal to the error threshold, calculate the second timestamp minus the first timestamp to obtain the startup time.

[0025] The beneficial effects brought by the technical solution provided in this application include:

[0026] This application identifies the actual longitudinal distance from the camera to the target object, subtracts it from the reference longitudinal distance to calculate the variance, and when the variance is less than the error threshold, obtains the startup time by calculating two timestamps. Since the two timestamps from the camera power-on to stable operation are accurately recorded, the obtained startup time of the camera is more accurate; and this application can be automatically performed in real time throughout the process without human participation, avoiding human errors, and making the detection efficiency and detection accuracy of the startup time higher.

[0027] The camera of this application is tested in a real environment or a simulation environment. The longitudinal distance is relatively more stable and accurate to identify compared with other parameters, such as the lateral distance, speed, or orientation angle, etc. Therefore, this application adopts the error threshold of the longitudinal distance from the camera to the target object. When the camera identifies the target object, the timestamps are calculated, making the obtained startup time of the intelligent camera more accurate and further accurately recording the startup time of the intelligent camera.

[0028] The technical solution of this application does not require cross-platform and heterogeneous platform synchronization processing, is simple and accurate to implement, can provide the precise startup time of the intelligent camera for the intelligent driving system integrator to confirm whether the startup time of the intelligent camera meets the integration requirements; at the same time, it can also be provided to the intelligent driving system diagnostic function development team for developing the intelligent camera fault diagnosis module. Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions in the embodiments of this application, the following will briefly introduce the drawings required for the description of the embodiments. Obviously, the following drawings are only some embodiments of this application. For those of ordinary skill in the art, without creative efforts, other drawings can be obtained based on these drawings.

[0030] Figure 1 It is the flowchart of the method for testing the startup time of the camera in the embodiment of this application;

[0031] Figure 2 It is the flowchart of the method for obtaining the reference longitudinal distance in the embodiment of this application. Detailed Embodiments

[0032] To make the objectives, technical solutions, and advantages of the embodiments of this application clearer, the following will clearly and completely describe the technical solutions in the embodiments of this application with reference to the accompanying drawings in the embodiments of this application. Obviously, the described embodiments are some, but not all, of the embodiments of this application. All other embodiments obtained by those of ordinary skill in the art based on the embodiments in this application without creative efforts shall fall within the scope of protection of this application.

[0033] The embodiments of this application provide a method for testing the startup time of an intelligent camera, which can solve the technical problem of low accuracy in obtaining the startup time of an intelligent camera in the related art, and realizes accurate recording of the startup time of the intelligent camera; the whole process can be carried out automatically and in real time without human participation, avoiding human errors, and the detection efficiency and detection accuracy of the startup time are higher; and this application does not require cross-platform and heterogeneous platform synchronization processing, and the implementation is simple and accurate.

[0034] As Figure 1 shown, an embodiment of a method for testing the startup time of a camera is provided, including the following steps:

[0035] S1: Set the target object.

[0036] S2: Obtain the reference longitudinal distance and error threshold from the camera to the target object.

[0037] S3: Power on the camera and record the first timestamp T1.

[0038] S4: Identify the actual longitudinal distance from the camera to the target object. After subtracting the reference longitudinal distance from the actual longitudinal distance, calculate the variance, and record the second timestamp T2 at the same time.

[0039] S5: Determine whether the variance is greater than the error threshold. If not, go to S6; if so, go to S7.

[0040] S6: Subtract the first timestamp from the second timestamp to obtain the startup time T, that is, T = T2 - T1, and end.

[0041] S7: The second timestamp T2 is invalid, and an error reporting process is performed.

[0042] In the above steps, both the reference longitudinal distance and the actual longitudinal distance belong to the longitudinal distance. The longitudinal distance refers to the distance from the target object in the direction of the camera optical axis and in front of the camera to the camera in a coordinate system with the camera as the origin. In the coordinate system with the camera as the origin, the coordinate origin is the optical center position of the camera, the X-axis and the Y-axis are respectively parallel to the X-axis and the Y-axis of the image coordinate system, and the Z-axis is the optical axis of the camera.

[0043] When the camera is tested in a simulation environment or a real environment, data such as longitudinal distance, lateral distance, speed, and orientation angle can be obtained; the above steps use the longitudinal distance data, and the longitudinal distance is more accurate in testing than other data, and the recognition is relatively more stable. Therefore, it is carried out by judging whether the longitudinal distance from the camera to the target object exceeds the threshold.

[0044] In the above step S1, setting the target object includes setting the longitudinal distance range between the target object and the camera, placing the target object within the longitudinal distance range, and the target object is directly facing the camera. Setting the target object also includes setting the type, position, or posture of the target object. The type of the target object should be within the recognizable range of the camera to be tested, and it should be a target with a clear outline and a distinct contrast with the background to ensure that the camera can easily recognize the target object.

[0045] Further, setting the longitudinal distance range between the target object and the camera includes: taking the principle that the longitudinal distance from the camera to the target object is not a null value, obtaining the minimum value and the maximum value of the longitudinal distance; the value range between the two is the longitudinal distance range.

[0046] In this embodiment, the longitudinal distance that is relatively easy to accurately obtain should be selected as the judgment parameter. In order to ensure the accuracy of obtaining the longitudinal distance data, the target object is preferably set within the range of 30 - 60 meters in front of the camera.

[0047] After setting the target object, in the above step S2, obtain the reference longitudinal distance from the camera to the target object. The reference longitudinal distance is obtained through the hardware-in-the-loop test (Hardware-in-the-Loop, abbreviated as HIL) of the camera, and is obtained by repeating the test, as Figure 2 shown, the specific steps are as follows:

[0048] S21: Set the number of tests.

[0049] S22: Power on the camera, and the camera performs a hardware-in-the-loop test.

[0050] S23: Obtain the longitudinal distance from the camera to the target object.

[0051] S24: Judge whether the longitudinal distance is null. If so, enter S25; if not, enter S21.

[0052] S25: Record the longitudinal distance, and power off the camera.

[0053] S26: Judge whether the set number of tests is reached. If not, enter S22; if so, enter S27.

[0054] S27: Calculate the mean value of all recorded longitudinal distances as the reference longitudinal distance.

[0055] In the above step S21, set the number of tests, that is, set the number of repeated tests N: To ensure that the finally obtained longitudinal distance data can be used for data validity determination and avoid excessive repetition affecting the test efficiency, the number of tests in this embodiment should be set to N = 5 - 10.

[0056] In the above step S24, judge the obtained longitudinal distance. If so, it means that the longitudinal distance is a null value, then it is determined that the camera does not recognize the current target object and the current test is unsuccessful.

[0057] In the above steps, if the number of tests is not reached, the camera needs to be powered on again, the test is repeated, and the longitudinal distance each time is recorded until the number of tests is reached and the test ends. Further, this repeated test is realized by building an automated test sequence to achieve fully automatic repeated power-on and power-off tests; after each power-off, after a preset time interval, the camera starts the next power-on.

[0058] In this embodiment, to ensure that the intelligent camera is initialized, it waits at least 60 seconds after power-off before it can be powered on again.

[0059] Further, in the process of obtaining the reference longitudinal distance from the camera to the target object, if the position, type or pose of the target object is changed and the longitudinal distance from the camera to the target object directly in front of the camera is not a null value, there is no need to obtain it again; otherwise, if the longitudinal distance is a null value and the position, type, pose or angle of the camera is changed, it needs to be obtained again.

[0060] In the above step S2, to obtain the error threshold, subtract the reference longitudinal distance from all the recorded longitudinal distances in turn, square them and then sum them up to get the variance, and this variance is used as the error threshold.

[0061] In the above step S4, the actual longitudinal distance recognized refers to that during the test, the camera obtains the longitudinal distance from the camera to the target object, and this longitudinal distance is not a null value, and based on this, it is judged that the camera recognizes the target object.

[0062] When the actual longitudinal distance is not a null value, subtract the actual longitudinal distance from the reference longitudinal distance, calculate the variance, and record T2 at the same time.

[0063] In the above steps S5 - S6, when the variance is less than or equal to the error threshold, the startup time T of the camera = T2 - T1.

[0064] In the above step S7, the error handling of the test is carried out in different cases: It may be that there is a problem with the target object setting, such as the longitudinal distance is too close or too far, which can be handled by resetting the target object; if the variance is still greater than the error threshold, it means that there is a problem with the parameter consistency of the camera, and it is handled by re-obtaining the reference longitudinal distance from the camera to the target object and the error threshold.

[0065] When obtaining the reference longitudinal distance and conducting the camera startup time test, ensure that the two are in exactly the same test scenario, which can be a real scenario or a simulation scenario.

[0066] This application also provides an embodiment of a camera startup time test system, which includes: an acquisition module, a timestamp recording module, and a calculation module.

[0067] The acquisition module is used to set the target object and obtain the reference longitudinal distance and error threshold of the camera.

[0068] The timestamp recording module is used to record the first timestamp T1 simultaneously when the camera is powered on; identify the actual longitudinal distance from the camera to the target object and record the second timestamp T2.

[0069] The calculation module is used to calculate the variance after subtracting the actual longitudinal distance from the reference longitudinal distance; when the variance is less than or equal to the error threshold, calculate the startup time T = T2 - T1.

[0070] The entire process of system testing for the camera startup time can be carried out in the same real-time system. The real-time system issues a power-on command, collects data, and analyzes the camera output data, ensuring that both power-on and data acquisition are through the same system, avoiding time asynchronization caused by cross-platform issues.

[0071] It should be noted that in this application, relational terms such as "first" and "second" are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not explicitly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the statement "including a..." does not exclude the existence of additional identical elements in the process, method, article or device including the said element.

[0072] The above are only specific embodiments of this application, enabling those skilled in the art to understand or implement this application. Various modifications to these embodiments will be obvious to those skilled in the art, and the general principles defined herein can be implemented in other embodiments without departing from the spirit or scope of this application. Therefore, this application will not be limited to these embodiments shown herein, but will conform to the widest scope consistent with the principles and novel features claimed herein.

Claims

1. A method for testing the startup time of an intelligent camera, where the startup time refers to the time interval from power-on to stable operation of the intelligent camera. Characterized in that, It includes: Set a target object, and obtain the reference longitudinal distance from the camera to the target object and the error threshold. Power on the camera, and record the first timestamp at the same time. Identify the actual longitudinal distance from the camera to the target object, calculate the variance after subtracting it from the reference longitudinal distance, and record the second timestamp at the same time. When the variance is less than or equal to the error threshold, subtract the first timestamp from the second timestamp to obtain the startup time.

2. The method for testing the startup time of an intelligent camera according to claim 1, Characterized in that, When the variance is greater than the error threshold, the second timestamp is invalid, and an error reporting process is performed.

3. The method for testing the startup time of an intelligent camera according to claim 1, Characterized in that, The setting of the target object includes: setting the longitudinal distance range between the target object and the camera, placing the target object within the longitudinal distance range, and the target object facing the camera directly.

4. The method for testing the startup time of an intelligent camera according to claim 3, Characterized in that, The setting of the longitudinal distance range between the target object and the camera includes: setting the target object, and obtaining the maximum value and the minimum value of the longitudinal distance on the principle that the longitudinal distance from the camera to the target object is not a null value; the value range between the two is the longitudinal distance range.

5. The method for testing the startup time of an intelligent camera according to claim 3, Characterized in that, Obtaining the reference longitudinal distance from the camera to the target object includes: Setting the number of tests, powering on the camera, performing hardware-in-the-loop testing on the camera, obtaining the longitudinal distance from the camera to the target object and recording it; Power off the camera and then power it on again, repeat the test and record the longitudinal distance each time until the number of tests is reached and the test ends; Calculate the mean value of all recorded longitudinal distances as the reference longitudinal distance.

6. The method for testing the startup time of an intelligent camera according to claim 5, Characterized in that, The repeated test is realized by building an automated test sequence to achieve fully automatic repeated power-on and power-off tests; after each power-off, after a preset time interval, the camera starts the next power-on.

7. The method for testing the startup time of an intelligent camera according to claim 5, Characterized in that, During the process of obtaining the reference longitudinal distance from the camera to the target object, if the position, type or posture of the target object is changed, and the longitudinal distance from the camera to the target object directly in front of the camera is not a null value, there is no need to obtain it again; Otherwise, if the longitudinal distance is a null value and the position, type, or posture of the camera is changed, it is necessary to obtain it again.

8. The method for testing the startup time of an intelligent camera according to claim 5, Characterized in that, Obtaining the error threshold includes: subtracting the reference longitudinal distance from all recorded longitudinal distances in turn, then squaring and summing them to obtain the variance, and taking this variance as the error threshold.

9. The method for testing the startup time of an intelligent camera according to any one of claims 1-8, It is characterized in that the longitudinal distance refers to the distance from the target in the positive front of the camera optical axis to the camera in the coordinate system with the camera as the origin.

10. An intelligent camera startup time test system, where the startup time refers to the time interval from when the intelligent camera is powered on to when it operates stably. It is characterized in that it includes: an acquisition module, configured to set a target and acquire the reference longitudinal distance and error threshold of the camera; a timestamp recording module, configured to record a first timestamp simultaneously when the camera is powered on; It is also configured to identify the actual longitudinal distance from the camera to the target and record a second timestamp; a calculation module, configured to calculate the variance after subtracting the actual longitudinal distance from the reference longitudinal distance; When the variance is less than or equal to the error threshold, calculate the second timestamp minus the first timestamp to obtain the startup time.

Images