Automobile camera image grayscale reproduction test method and system

Abstract

The embodiment of the present invention provides a method and system for grayscale reproduction testing of automotive camera images, which belongs to the field of grayscale testing technology. The method includes: recovering the scene image of the current scene captured by the camera to be tested on the vehicle side; parsing the scene image, calibrating the parameter information of the camera to be tested based on the parsing result, and obtaining calibration parameters; performing experimental camera parameter adjustment, and performing grayscale card shooting based on the experimental camera and card shooting device after parameter adjustment; obtaining the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the grayscale card shooting image corresponding to the actual parameters and the grayscale card shooting image corresponding to the calibration parameters respectively; generating the grayscale reproduction result of the camera to be tested based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters. The scheme of the present invention not only improves the efficiency of grayscale testing, but also improves the accuracy of grayscale testing.

Description

Technical Field

[0001] The present invention relates to the technical field of grayscale testing, and in particular to a grayscale level reproduction testing method for an automobile camera image and a grayscale level reproduction testing system for an automobile camera image. Background Art

[0002] CMS (Camera Monitor System) is a monitoring system that uses the images captured by the camera on the vehicle display, such as the smart rearview mirror. The CMS camera monitoring system is divided into three necessary functional modules according to its functions: image capture module, image processing module, and image display module. In order to ensure driving safety, the CMS system has very high requirements for pre-delivery testing. During the use of the vehicle, there are also problems such as imaging anomalies caused by changes in the distortion parameters of the vehicle camera that require self-testing and resetting. In response to this usage requirement, there are corresponding imaging effect tests, including grayscale reproduction tests.

[0003] Grayscale reproduction test involves grayscale test. Grayscale test is to create a scene with obvious brightness level relationship from darkest black to brightest white, and record the camera's response to the light and dark contrast and transition from white to black in this scene. The clearer the grayscale level in the captured picture, the more natural the transition between adjacent grayscales, the better the grayscale response ability of the camera, and the more satisfactory the black and white contrast. In the existing test method, it is necessary to calculate the brightness value for each grayscale color block, and then convert the result based on the calculated value. In this case, it is necessary to ensure that the display performance of the screen meets the standard and configure the corresponding color space to achieve accurate calculation. The calculation error of the brightness value of any grayscale color block will lead to the error of the test result. And it is necessary to calculate the brightness value for each grayscale color block, resulting in a large amount of calculation and a high test delay. In view of the problems of low accuracy and low test efficiency in the existing grayscale reproduction test scheme of automotive camera images, a new test scheme needs to be proposed. Summary of the invention

[0004] The purpose of the embodiments of the present invention is to provide a method and system for reproducing grayscale levels of automobile camera images, so as to at least solve the problems of low accuracy and low test efficiency in existing automobile camera image grayscale level reproduction test schemes.

[0005] In order to achieve the above-mentioned purpose, the first aspect of the present invention provides a grayscale reproduction test method for an automobile camera image, the method being executed by a self-test end, the self-test end including an experimental camera and a picture card shooting device, the method including: recovering a scene image of the current scene captured by the camera to be tested at the vehicle end; parsing the scene image, calibrating the parameter information of the camera to be tested based on the parsing result, and obtaining the calibration parameters; sequentially performing experimental camera parameter adjustment based on the actual parameters and calibration parameters of the camera to be tested, and performing grayscale picture card shooting based on the adjusted experimental camera and picture card shooting device; calculating the grayscale brightness ratio based on the grayscale picture card shot image corresponding to the actual parameters and the grayscale picture card shot image corresponding to the calibration parameters, respectively, and obtaining the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the calculation results; generating the grayscale reproduction result of the camera to be tested based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters.

[0006] Optionally, the picture card shooting device includes: a neutral gray backboard for placing a grayscale picture card; an illumination cavity with the central gray backboard as the bottom surface of the cone; the illumination cavity is composed of a light-blocking baffle, and a fill light source is arranged on the light-blocking baffle; a shooting hole is arranged at the vertex position of the cone formed by the illumination cavity and the central gray backboard, which is used to set an experimental camera.

[0007] Optionally, parsing the scene image and calibrating parameter information of the camera to be tested based on the parsing result to obtain calibration parameters includes: calculating the horizontal field of view of the camera to be tested based on the scene image to obtain a calculated value of the horizontal field of view, and the calculation rule is:

[0008]

[0009] Wherein, f is the focal length information of the scene image; w is the width of the scene image; H is the horizontal field of view information of the camera to be tested; the distortion parameters of the camera to be tested are corrected based on the calculated value of the horizontal field of view and the actual value of the horizontal field of view corresponding to the actual parameters of the camera to be tested; and the parameter information corresponding to the camera to be tested after the distortion parameter correction is used as the calibration parameter.

[0010] Optionally, the experimental camera parameters are adjusted in sequence based on the actual parameters and calibration parameters of the camera to be tested, and grayscale card shooting is performed based on the adjusted experimental camera and the card shooting device, including: adjusting the experimental camera parameters in sequence based on the actual parameters and calibration parameters of the camera to be tested, and after completing one parameter adjustment, grayscale card shooting is performed based on the experimental camera and the card shooting device to obtain grayscale card shooting images corresponding to the actual parameters and grayscale card shooting images corresponding to the calibration parameters.

[0011] Optionally, the grayscale brightness ratio is calculated based on the grayscale card image corresponding to the actual parameters and the grayscale card image corresponding to the calibration parameters, and the grayscale level reproduction result of the actual parameters and the grayscale level reproduction result under the calibration parameters are obtained based on the calculation results, including: in each captured image, each grayscale color block is identified based on an adaptive threshold segmentation algorithm; the grayscale value of each grayscale color block is identified, and each grayscale color block is assigned a value based on the grayscale value; whether the assignment result of each continuous grayscale color block is a continuous value, and if not, re-shooting the grayscale image; The step chart is used until the assignment result of the continuous grayscale color blocks is a continuous value; the brightness value of each grayscale color block is identified, and based on the brightness value of each grayscale color block, the brightness ratio of each grayscale color block is calculated; based on the brightness ratio of each grayscale color block, the brightness difference between any two adjacent grayscale color blocks is calculated to obtain a brightness difference set of actual parameters and a brightness difference set of calibration parameters respectively; based on the brightness difference set of actual parameters and the brightness difference set of calibration parameters, the grayscale level reproduction result of the actual parameters and the grayscale level reproduction result under the calibration parameters are determined respectively.

[0012] Optionally, assigning values ​​to the grayscale color blocks based on the grayscale values ​​includes: sorting the grayscale color blocks based on their grayscale values, with the higher the grayscale value, the higher the sorting; and assigning values ​​to the grayscale color blocks based on the sorting result.

[0013] Optionally, the identifying the brightness value of each grayscale color block and calculating the brightness ratio of each grayscale color block based on the brightness value of each grayscale color block include: calculating the brightness ratio between any two adjacent grayscale color blocks based on the brightness value of each grayscale color block, traversing all grayscale color blocks to obtain a brightness ratio set; converting to obtain a corresponding brightness ratio set based on the brightness ratio set; calculating the brightness difference between any two adjacent grayscale color blocks based on the brightness ratio of each grayscale color block, and respectively obtaining a brightness difference set of actual parameters and a brightness difference set of calibration parameters, including: assigning brightness to the grayscale color block assigned a value of 1, and assigning brightness to each grayscale color block in turn based on the brightness assignment of the grayscale color block assigned a value of 1 and the brightness ratio set to obtain a brightness assignment result of each grayscale color block; based on the brightness assignment result of each grayscale color block, settling the brightness difference between each two adjacent grayscale color blocks in turn to obtain a brightness difference set.

[0014] Optionally, the determining of the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters respectively includes: traversing the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters, respectively comparing each brightness difference with a preset difference threshold, and respectively recording the number of brightness difference values ​​corresponding to the actual parameters that are greater than the preset difference threshold and the number of brightness difference values ​​corresponding to the calibration parameters that are greater than the preset difference threshold, as the first brightness difference number and the second brightness difference number, respectively; if the first brightness difference number is greater than the preset difference number threshold, it indicates that the grayscale reproduction result of the actual parameters is qualified, otherwise, it is unqualified; if the second brightness difference number is greater than the preset difference number threshold, it indicates that the grayscale reproduction result under the calibration parameters is qualified, otherwise, it is unqualified; the determining of the grayscale reproduction result based on the actual parameters The grayscale reproduction result and the grayscale reproduction result under the calibration parameters generate the grayscale reproduction result of the camera to be tested, including: if the grayscale reproduction result of the actual parameters is unqualified, but the grayscale reproduction result under the calibration parameters is qualified, it means that the test is invalid, and the scene image is re-collected to perform the test; if the grayscale reproduction result of the actual parameters is qualified, and the grayscale reproduction result under the calibration parameters is qualified, it means that the grayscale reproduction result of the camera to be tested is qualified; if the grayscale reproduction result of the actual parameters is unqualified, and the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified; if the grayscale reproduction result of the actual parameters is qualified, but the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified, and the camera to be tested is adjusted based on the calibration parameters.

[0015] The second aspect of the present invention provides a grayscale reproduction test system for an automobile camera image, the system comprising a vehicle-mounted end and a self-test end that are communicatively connected to each other; the camera to be tested is arranged at the vehicle-mounted end, and the self-test end comprises: an experimental camera and a picture card shooting device; an acquisition unit, used to recover the scene image of the current scene acquired by the camera to be tested at the vehicle-mounted end; a calibration unit, used to parse the scene image, calibrate the parameter information of the camera to be tested based on the analysis result, and obtain the calibration parameters; a testing unit, used to perform experimental camera parameter adjustment in sequence based on the actual parameters and calibration parameters of the camera to be tested, and perform grayscale picture card shooting based on the adjusted experimental camera and picture card shooting device; a processing unit, used to calculate the grayscale brightness ratio based on the grayscale picture card shooting image corresponding to the actual parameters and the grayscale picture card shooting image corresponding to the calibration parameters, respectively, and obtain the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the calculation results; a determination unit, used to generate the grayscale reproduction result of the camera to be tested based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters.

[0016] On the other hand, the present invention provides a computer-readable storage medium having instructions stored thereon, which, when executed on a computer, enables the computer to execute the above-mentioned automobile camera image grayscale reproduction test method.

[0017] Through the above technical scheme, the scheme of the present invention simulates the camera to be tested at the self-test end to adjust the experimental camera parameters based on the scene image transmitted back from the vehicle end, and then performs a self-test test based on the experimental camera, without directly testing the vehicle end, thereby realizing remote testing of the vehicle and improving test efficiency. The scheme of the present invention does not need to display the captured image, nor does it need to calculate the brightness of the grayscale color block. Instead, it is directly based on the brightness ratio of the grayscale color block, thereby limiting other conditions through the chart shooting device to directly convert the brightness ratio, and determine the brightness difference through the brightness ratio, thereby evaluating the camera's accuracy in distinguishing grayscale color blocks, thereby realizing a reproducible test, no longer relying on the colorimetric system and display device, and ensuring test accuracy.

[0018] Other features and advantages of the embodiments of the present invention will be described in detail in the following detailed description. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific embodiments, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0020] Figure 1 It is a flowchart of the steps of a grayscale reproduction test method for an automobile camera image provided by an embodiment of the present invention;

[0021] Figure 2 is a result schematic diagram of a picture card shooting device provided by an embodiment of the present invention;

[0022] Figure 3 The present invention is a schematic diagram of the structure of a self-test terminal in a grayscale reproduction test system for an automobile camera image provided by an embodiment of the present invention.

[0023] Description of Reference Numerals

[0024] 10- neutral grey back panel; 20- light blocking baffle; 30- fill light source; 40- shooting hole; 50- experimental camera. DETAILED DESCRIPTION

[0025] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0026] Figure 11 is a flow chart of a method for reproducing grayscale levels of an automobile camera image provided by an embodiment of the present invention. Figure 1 As shown, an embodiment of the present invention provides a method for reproducing grayscale levels of an automobile camera image, the method comprising:

[0027] Step S10: Recover the scene image of the current scene captured by the camera to be tested at the vehicle terminal.

[0028] Specifically, the scheme of the present invention is used to test the vehicle CMS system, and the corresponding test object must be the car camera. In the actual application process, it is necessary to test the camera itself. If the vehicle itself is still in the pre-factory stage, it can be convenient to carry out centralized testing. However, during the use of the vehicle, as the use time increases, the internal parameters of the vehicle camera will change to a certain extent, resulting in changes in image distortion. In order to ensure driving safety, it is of practical significance to adjust the corresponding parameters of this distortion. In this case, it is necessary to self-check and adjust the camera of the vehicle in use. However, if the vehicle needs to be driven to a designated location for testing, it is bound to lead to lower test efficiency and reduced user experience. Based on this, the scheme of the present invention proposes an online self-checking scheme, which can realize remote self-checking by transmitting the scene image collected by the vehicle terminal to the self-checking terminal. Based on this, the scheme of the present invention will collect scene images of the corresponding position through the camera to be tested at the extreme of the vehicle, so as to facilitate subsequent testing of the scene image.

[0029] Step S20: Analyze the scene image, and calibrate the parameter information of the camera to be tested based on the analysis result to obtain calibration parameters.

[0030] Specifically, the horizontal field of view of the camera to be tested is calculated based on the scene image to obtain a calculated value of the horizontal field of view. The calculation rule is:

[0031]

[0032] Wherein, f is the focal length information of the scene image; w is the width of the scene image; H is the horizontal field of view information of the camera to be tested; the distortion parameters of the camera to be tested are corrected based on the calculated value of the horizontal field of view and the actual value of the horizontal field of view corresponding to the actual parameters of the camera to be tested; and the parameter information corresponding to the camera to be tested after the distortion parameter correction is used as the calibration parameter.

[0033] In an embodiment of the present invention, after the solution of the present invention recovers the scene image collected by the camera to be tested, it is necessary to calibrate the internal parameters of the current camera based on the collected scene image to determine whether the experimental camera parameter adjustment directly based on the current camera can restore the actual situation of the camera to be tested. Because the parameters read by the camera to be tested may have errors with its actual situation (internal parameter deviation caused by the increase in usage time), in order to ensure that the simulation situation is consistent with the actual situation of the current camera, it is necessary to calibrate the actual internal parameters, so as to facilitate subsequent self-test shooting based on the calibrated internal parameters to ensure test accuracy. Based on this, the solution of the present invention determines the distortion situation based on the horizontal field of view in the internal parameters, calculates the actual horizontal field of view by collecting images, and compares it with the designed horizontal field of view, so as to obtain the corresponding distortion deviation parameters, and adjusts the experimental camera parameters based on the distortion deviation parameters, which can ensure that the simulation situation matches the actual situation.

[0034] Step S30: perform parameter adjustment of the experimental camera in sequence based on the actual parameters and calibration parameters of the camera to be tested, and perform grayscale card shooting based on the adjusted experimental camera and card shooting device.

[0035] Specifically, the experimental camera is adjusted based on the actual parameters and calibration parameters of the camera to be tested. After completing one adjustment, grayscale card shooting is performed based on the experimental camera and the card shooting device to obtain grayscale card shooting images corresponding to the actual parameters and grayscale card shooting images corresponding to the calibration parameters.

[0036] In the embodiment of the present invention, if the parameters are adjusted only based on the simulated distortion, it can only determine whether the grayscale reproduction test of the camera of the current vehicle is qualified, and the parameters of the vehicle camera cannot be adjusted based on the current test results, so that the vehicle camera can overcome the problem of image distortion based on parameter adjustment. Based on this, the scheme of the present invention also wants to achieve that the vehicle terminal can adjust the camera parameters based on the test results, so that the vehicle camera can alleviate the problem of image distortion through parameter adjustment, thereby ensuring the imaging effect, so as to ensure higher driving safety.

[0037] Based on this, in addition to adjusting the experimental camera parameters based on the calibrated camera parameters, the solution of the present invention will also perform a set of experimental camera parameter adjustments based on the actual parameters of the camera to be tested, so as to obtain the actual grayscale level reproduction test results under the vehicle camera design parameters, thereby determining the deviation value of the vehicle camera's current situation from the design situation, and subsequently adjusting the vehicle camera parameters based on the deviation value.

[0038] Preferably, Figure 2The image card shooting device includes: a neutral gray back plate for placing a grayscale image card; an illumination cavity with the central gray back plate as the bottom surface of the cone; the illumination cavity is composed of a light-blocking baffle, and a fill light source is arranged on the light-blocking baffle; a shooting hole is arranged at the vertex position of the cone formed by the illumination cavity and the central gray back plate, which is used to set the experimental camera.

[0039] In one possible implementation, when shooting a grayscale color card, adsorb the grayscale test card on a neutral gray backplane, center it left and right, and level it horizontally. Adjust the color temperature and brightness level of the fill light source to the target value (6500K: 1500Lux), and ensure that the surface uniformity of the test card is controlled within ±5%. Adjust the relative position of the experimental camera and the test card until the 12 grayscale blocks of the grayscale card are displayed in the center of the display area defined by the experimental monitor, and each grayscale block image displays at least 50×50 pixels. Check whether the equipment lens is dirty. If so, wipe it with a lens cloth. If the lens is damaged, replace the equipment. Level the tripod and gimbal, adjust the picture, and shoot the image in a moderate ratio to ensure that the grayscale area accounts for more than 1000 pixels, and the surface of the test card in the picture is not reflective; the lens surface and the test card surface are relatively parallel and the picture is symmetrical up and down and left and right. When shooting, first block the lens, re-trigger 3A, and shoot after the picture is stable. It is recommended to shoot 3 sample images.

[0040] Step S40: Calculate the grayscale brightness ratio based on the grayscale chart image corresponding to the actual parameters and the grayscale chart image corresponding to the calibration parameters, and obtain the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the calculation results.

[0041] Specifically, in each captured image, each grayscale color block is identified based on an adaptive threshold segmentation algorithm; the grayscale value of each grayscale color block is identified, and each grayscale color block is assigned a value based on the grayscale value; it is determined whether the assignment result of each continuous grayscale color block is a continuous value, if not, the grayscale chart is re-shot until the assignment result of the continuous grayscale color block is a continuous value; the brightness value of each grayscale color block is identified, and based on the brightness value of each grayscale color block, the brightness ratio of each grayscale color block is calculated; based on the brightness ratio of each grayscale color block, the brightness difference between any two adjacent grayscale color blocks is calculated, and the brightness difference set of actual parameters and the brightness difference set of calibration parameters are respectively obtained; based on the brightness difference set of actual parameters and the brightness difference set of calibration parameters, the grayscale level reproduction result of the actual parameters and the grayscale level reproduction result under the calibration parameters are respectively determined.

[0042] In an embodiment of the present invention, in order to ensure the accuracy of the test results, it is necessary to conduct a quality audit on the collected color card image, because there may still be accidental errors that lead to abnormal image shooting. If the abnormal image is directly processed to obtain the test result, an erroneous test result will be obtained. In order to avoid this situation, after completing the image acquisition, the solution of the present invention needs to perform contour recognition of each grayscale color block based on the recognition result, and then perform grayscale recognition of each area under each recognized contour to label each grayscale color block based on the recognition result. If a shooting abnormality occurs, such as a shadow, the serial number of the identification area corresponding to the shadow will be out of order based on the randomness of the shadow. The solution of the present invention compares the number of serial numbers of each area and the continuity of the serial numbers to determine whether there is an accidental shadow area, thereby avoiding the problem of shooting abnormalities.

[0043] Furthermore, assigning values ​​to the grayscale color blocks based on the grayscale values ​​includes: sorting the grayscale color blocks based on their grayscale values, with the higher the grayscale value, the higher the sorting; and assigning values ​​to the grayscale color blocks based on the sorting result.

[0044] Further, the identifying the brightness value of each grayscale color block and calculating the brightness ratio of each grayscale color block based on the brightness value of each grayscale color block include: calculating the brightness ratio between any two adjacent grayscale color blocks based on the brightness value of each grayscale color block, traversing all grayscale color blocks to obtain a brightness ratio set; converting to obtain a corresponding brightness ratio set based on the brightness ratio set; calculating the brightness difference between any two adjacent grayscale color blocks based on the brightness ratio of each grayscale color block, and respectively obtaining a brightness difference set of actual parameters and a brightness difference set of calibration parameters, including: assigning brightness to the grayscale color block assigned a value of 1, and assigning brightness to each grayscale color block in turn based on the brightness assignment of the grayscale color block assigned a value of 1 and the brightness ratio set to obtain a brightness assignment result of each grayscale color block; based on the brightness assignment result of each grayscale color block, settling the brightness difference between each two adjacent grayscale color blocks in turn to obtain a brightness difference set.

[0045] In the embodiment of the present invention, it has been explained above that the grayscale test is to create a scene with a clear brightness level relationship from the darkest black to the brightest white, and record the camera's response to the light and dark contrast and the transition from white to black in this scene. The clearer the grayscale levels in the captured image, the more natural the transition between adjacent grayscales, and the better the grayscale response ability of the camera. Based on this, it is only necessary to determine whether the chromaticity difference between two adjacent grayscale color blocks is large enough (the clearer the corresponding level), and then the corresponding grayscale reproduction ability can be understood. Based on this, the scheme of the present invention only needs to preset the initial value based on the brightness ratio, and the brightness of the subsequent color blocks can be directly assigned based on the preset value. Through the numerical difference between the assigned values, the difference between the brightness values ​​can be equivalently replaced.

[0046] Preferably, the method of determining the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters respectively includes: traversing the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters, respectively comparing each brightness difference with a preset difference threshold, and respectively recording the number of brightness difference values ​​corresponding to the actual parameters that are greater than the preset difference threshold and the number of brightness difference values ​​corresponding to the calibration parameters that are greater than the preset difference threshold, as the first brightness difference number and the second brightness difference number, respectively; if the first brightness difference number is greater than the preset difference number threshold, it indicates that the grayscale reproduction result of the actual parameters is qualified, otherwise, it is unqualified; if the second brightness difference number is greater than the preset difference number threshold, it indicates that the grayscale reproduction result under the calibration parameters is qualified, otherwise, it is unqualified.

[0047] Furthermore, the brightness value of the color block can be directly read, and other external factors are limited by the image capture device, resulting in that the ratio of the brightness value can be directly equivalent to the ratio of the brightness value (other variables are exactly the same). Based on this, the scheme of the present invention obtains the brightness ratio of the color block by reading the brightness value of each grayscale color block, so as to convert the brightness difference based on the brightness ratio. Through the assignment scheme, the corresponding difference threshold is preset. When it is greater than the difference threshold, it means that the obvious degree of differentiation level meets the expectation. If it is less than the threshold, it means that the differentiation level between the two color blocks is not obvious, so it does not meet the expectation. By counting the number of differences greater than the difference threshold, the color blocks in the test result that meet the expected differentiation effect can be known. Then, by presetting the difference number threshold, it can be determined whether the test result meets the expectation. For example, if the target color card has 12 color blocks, there are 11 differences, and the preset difference threshold is 8. If 9 or more of the 11 differences are greater than the preset difference threshold, it means that the test result is qualified, otherwise, it is judged as unqualified.

[0048] Step S50: generating a grayscale reproduction result of the camera to be tested based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters.

[0049] Specifically, if the grayscale reproduction result of the actual parameters is unqualified, but the grayscale reproduction result under the calibration parameters is qualified, it means that the test is invalid and the scene image is re-captured to perform the test; if the grayscale reproduction result of the actual parameters is qualified, and the grayscale reproduction result under the calibration parameters is qualified, it means that the grayscale reproduction result of the camera to be tested is qualified; if the grayscale reproduction result of the actual parameters is unqualified, and the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified; if the grayscale reproduction result of the actual parameters is qualified, but the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified, and the camera to be tested is adjusted based on the calibration parameters.

[0050] In an embodiment of the present invention, in theory, the design effect is better than the actual effect after long-term use. Based on this, if the test results corresponding to the actual parameters are unqualified, but the test results of the calibration parameters are qualified, it means that the actual effect after long-term use is better than the design effect, which is contrary to the theoretical situation, indicating that there is a test error and it is necessary to re-collect images for testing. If both are qualified, it means that the vehicle camera test is qualified and no parameter adjustment is required. If both are unqualified, it means that the product is abnormal and corresponding maintenance and replacement operations are required. If the test results corresponding to the actual parameters are qualified, but the test results of the calibration parameters are unqualified, it means that the design effect of the vehicle camera is in line with expectations, but because of the parameter deviation caused by long-term use, the parameters need to be adjusted to ensure that the effect is in line with expectations, then the parameters of the camera to be tested are adjusted based on the calibration parameters.

[0051] Figure 3 FIG. 1 is a system structure diagram of a grayscale reproduction test system for an automobile camera image provided by an embodiment of the present invention. Figure 3 As shown, an embodiment of the present invention provides a car camera image grayscale reproduction test system, the system includes a vehicle end and a self-test end that are communicatively connected to each other; the camera to be tested is arranged at the vehicle end, and the self-test end includes: an experimental camera and a picture card shooting device.

[0052] The acquisition unit is used to recover the scene image of the current scene acquired by the camera to be tested on the vehicle side.

[0053] Specifically, the scheme of the present invention is used to test the vehicle CMS system, and the corresponding test object must be the car camera. In the actual application process, it is necessary to test the camera itself. If the vehicle itself is still in the pre-factory stage, it can be convenient to carry out centralized testing. However, during the use of the vehicle, as the use time increases, the internal parameters of the vehicle camera will change to a certain extent, resulting in changes in image distortion. In order to ensure driving safety, it is of practical significance to adjust the corresponding parameters of this distortion. In this case, it is necessary to self-check and adjust the camera of the vehicle in use. However, if the vehicle needs to be driven to a designated location for testing, it is bound to lead to lower test efficiency and reduced user experience. Based on this, the scheme of the present invention proposes an online self-checking scheme, which can realize remote self-checking by transmitting the scene image collected by the vehicle terminal to the self-checking terminal. Based on this, the scheme of the present invention will collect scene images of the corresponding position through the camera to be tested at the extreme of the vehicle, so as to facilitate subsequent testing of the scene image.

[0054] The calibration unit is used to analyze the scene image, and calibrate the parameter information of the camera to be tested based on the analysis result to obtain calibration parameters.

[0055] Specifically, the horizontal field of view of the camera to be tested is calculated based on the scene image to obtain a calculated value of the horizontal field of view. The calculation rule is:

[0056]

[0057] Wherein, f is the focal length information of the scene image; w is the width of the scene image; H is the horizontal field of view information of the camera to be tested; the distortion parameters of the camera to be tested are corrected based on the calculated value of the horizontal field of view and the actual value of the horizontal field of view corresponding to the actual parameters of the camera to be tested; and the parameter information corresponding to the camera to be tested after the distortion parameter correction is used as the calibration parameter.

[0058] In an embodiment of the present invention, after the solution of the present invention recovers the scene image collected by the camera to be tested, it is necessary to calibrate the internal parameters of the current camera based on the collected scene image to determine whether the experimental camera parameter adjustment directly based on the current camera can restore the actual situation of the camera to be tested. Because the parameters read by the camera to be tested may have errors with its actual situation (internal parameter deviation caused by the increase in usage time), in order to ensure that the simulation situation is consistent with the actual situation of the current camera, it is necessary to calibrate the actual internal parameters, so as to facilitate subsequent self-test shooting based on the calibrated internal parameters to ensure test accuracy. Based on this, the solution of the present invention determines the distortion situation based on the horizontal field of view in the internal parameters, calculates the actual horizontal field of view by collecting images, and compares it with the designed horizontal field of view, so as to obtain the corresponding distortion deviation parameters, and adjusts the experimental camera parameters based on the distortion deviation parameters, which can ensure that the simulation situation matches the actual situation.

[0059] The test unit is used to perform experimental camera parameter adjustment in sequence based on the actual parameters and calibration parameters of the camera to be tested, and to perform grayscale picture card shooting based on the adjusted experimental camera and picture card shooting device.

[0060] Specifically, the experimental camera is adjusted based on the actual parameters and calibration parameters of the camera to be tested. After completing one adjustment, grayscale card shooting is performed based on the experimental camera and the card shooting device to obtain grayscale card shooting images corresponding to the actual parameters and grayscale card shooting images corresponding to the calibration parameters.

[0061] In the embodiment of the present invention, if the parameters are adjusted only based on the simulated distortion, it can only determine whether the grayscale reproduction test of the camera of the current vehicle is qualified, and the parameters of the vehicle camera cannot be adjusted based on the current test results, so that the vehicle camera can overcome the problem of image distortion based on parameter adjustment. Based on this, the scheme of the present invention also wants to achieve that the vehicle terminal can adjust the camera parameters based on the test results, so that the vehicle camera can alleviate the problem of image distortion through parameter adjustment, thereby ensuring the imaging effect, so as to ensure higher driving safety.

[0062] Based on this, in addition to adjusting the experimental camera parameters based on the calibrated camera parameters, the solution of the present invention will also perform a set of experimental camera parameter adjustments based on the actual parameters of the camera to be tested, so as to obtain the actual grayscale level reproduction test results under the vehicle camera design parameters, thereby determining the deviation value of the vehicle camera's current situation from the design situation, and subsequently adjusting the vehicle camera parameters based on the deviation value.

[0063] The processing unit is used to calculate the grayscale brightness ratio based on the grayscale chart image corresponding to the actual parameters and the grayscale chart image corresponding to the calibration parameters, and obtain the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the calculation results.

[0064] Specifically, in each captured image, each grayscale color block is identified based on an adaptive threshold segmentation algorithm; the grayscale value of each grayscale color block is identified, and each grayscale color block is assigned a value based on the grayscale value; it is determined whether the assignment result of each continuous grayscale color block is a continuous value, if not, the grayscale chart is re-shot until the assignment result of the continuous grayscale color block is a continuous value; the brightness value of each grayscale color block is identified, and based on the brightness value of each grayscale color block, the brightness ratio of each grayscale color block is calculated; based on the brightness ratio of each grayscale color block, the brightness difference between any two adjacent grayscale color blocks is calculated, and the brightness difference set of actual parameters and the brightness difference set of calibration parameters are respectively obtained; based on the brightness difference set of actual parameters and the brightness difference set of calibration parameters, the grayscale level reproduction result of the actual parameters and the grayscale level reproduction result under the calibration parameters are respectively determined.

[0065] In an embodiment of the present invention, in order to ensure the accuracy of the test results, it is necessary to conduct a quality audit on the collected color card image, because there may still be accidental errors that lead to abnormal image shooting. If the abnormal image is directly processed to obtain the test result, an erroneous test result will be obtained. In order to avoid this situation, after completing the image acquisition, the solution of the present invention needs to perform contour recognition of each grayscale color block based on the recognition result, and then perform grayscale recognition of each area under each recognized contour to label each grayscale color block based on the recognition result. If a shooting abnormality occurs, such as a shadow, the serial number of the identification area corresponding to the shadow will be out of order based on the randomness of the shadow. The solution of the present invention compares the number of serial numbers of each area and the continuity of the serial numbers to determine whether there is an accidental shadow area, thereby avoiding the problem of shooting abnormalities.

[0066] Furthermore, assigning values ​​to the grayscale color blocks based on the grayscale values ​​includes: sorting the grayscale color blocks based on their grayscale values, with the higher the grayscale value, the higher the sorting; and assigning values ​​to the grayscale color blocks based on the sorting result.

[0067] Further, the identifying the brightness value of each grayscale color block and calculating the brightness ratio of each grayscale color block based on the brightness value of each grayscale color block include: calculating the brightness ratio between any two adjacent grayscale color blocks based on the brightness value of each grayscale color block, traversing all grayscale color blocks to obtain a brightness ratio set; converting to obtain a corresponding brightness ratio set based on the brightness ratio set; calculating the brightness difference between any two adjacent grayscale color blocks based on the brightness ratio of each grayscale color block, and respectively obtaining a brightness difference set of actual parameters and a brightness difference set of calibration parameters, including: assigning brightness to the grayscale color block assigned a value of 1, and assigning brightness to each grayscale color block in turn based on the brightness assignment of the grayscale color block assigned a value of 1 and the brightness ratio set to obtain a brightness assignment result of each grayscale color block; based on the brightness assignment result of each grayscale color block, settling the brightness difference between each two adjacent grayscale color blocks in turn to obtain a brightness difference set.

[0068] In the embodiment of the present invention, it has been explained above that the grayscale test is to create a scene with a clear brightness level relationship from the darkest black to the brightest white, and record the camera's response to the light and dark contrast and the transition from white to black in this scene. The clearer the grayscale levels in the captured image, the more natural the transition between adjacent grayscales, and the better the grayscale response ability of the camera. Based on this, it is only necessary to determine whether the chromaticity difference between two adjacent grayscale color blocks is large enough (the clearer the corresponding level), and then the corresponding grayscale reproduction ability can be understood. Based on this, the scheme of the present invention only needs to preset the initial value based on the brightness ratio, and the brightness of the subsequent color blocks can be directly assigned based on the preset value. Through the numerical difference between the assigned values, the difference between the brightness values ​​can be equivalently replaced.

[0069] Preferably, the method of determining the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters respectively includes: traversing the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters, respectively comparing each brightness difference with a preset difference threshold, and respectively recording the number of brightness difference values ​​corresponding to the actual parameters that are greater than the preset difference threshold and the number of brightness difference values ​​corresponding to the calibration parameters that are greater than the preset difference threshold, as the first brightness difference number and the second brightness difference number, respectively; if the first brightness difference number is greater than the preset difference number threshold, it indicates that the grayscale reproduction result of the actual parameters is qualified, otherwise, it is unqualified; if the second brightness difference number is greater than the preset difference number threshold, it indicates that the grayscale reproduction result under the calibration parameters is qualified, otherwise, it is unqualified.

[0070] Furthermore, the brightness value of the color block can be directly read, and other external factors are limited by the image capture device, resulting in that the ratio of the brightness value can be directly equivalent to the ratio of the brightness value (other variables are exactly the same). Based on this, the scheme of the present invention obtains the brightness ratio of the color block by reading the brightness value of each grayscale color block, so as to convert the brightness difference based on the brightness ratio. Through the assignment scheme, the corresponding difference threshold is preset. When it is greater than the difference threshold, it means that the obvious degree of differentiation level meets the expectation. If it is less than the threshold, it means that the differentiation level between the two color blocks is not obvious, so it does not meet the expectation. By counting the number of differences greater than the difference threshold, the color blocks in the test result that meet the expected differentiation effect can be known. Then, by presetting the difference number threshold, it can be determined whether the test result meets the expectation. For example, if the target color card has 12 color blocks, there are 11 differences, and the preset difference threshold is 8. If 9 or more of the 11 differences are greater than the preset difference threshold, it means that the test result is qualified, otherwise, it is judged as unqualified.

[0071] The determination unit is used to generate a grayscale reproduction result of the camera to be tested based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters.

[0072] Specifically, if the grayscale reproduction result of the actual parameters is qualified, but the grayscale reproduction result under the calibration parameters is qualified or unqualified, it means that the test is invalid and the scene image is recaptured to perform the test; if the grayscale reproduction result of the actual parameters is qualified and the grayscale reproduction result under the calibration parameters is qualified, it means that the grayscale reproduction result of the camera to be tested is qualified; if the grayscale reproduction result of the actual parameters is unqualified and the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified; if the grayscale reproduction result of the actual parameters is unqualified, but the grayscale reproduction result under the calibration parameters is qualified, it means that the grayscale reproduction result of the camera to be tested is unqualified, and the camera to be tested is adjusted based on the calibration parameters.

[0073] In an embodiment of the present invention, in theory, the design effect is better than the actual effect after long-term use. Based on this, if the test results corresponding to the actual parameters are unqualified, but the test results of the calibration parameters are qualified, it means that the actual effect after long-term use is better than the design effect, which is contrary to the theoretical situation, indicating that there is a test error and it is necessary to re-collect images for testing. If both are qualified, it means that the vehicle camera test is qualified and no parameter adjustment is required. If both are unqualified, it means that the product is abnormal and corresponding maintenance and replacement operations are required. If the test results corresponding to the actual parameters are qualified, but the test results of the calibration parameters are unqualified, it means that the design effect of the vehicle camera is in line with expectations, but because of the parameter deviation caused by long-term use, the parameters need to be adjusted to ensure that the effect is in line with expectations, then the parameters of the camera to be tested are adjusted based on the calibration parameters.

[0074] An embodiment of the present invention further provides a computer-readable storage medium, on which instructions are stored, which, when executed on a computer, enable the computer to execute the above-mentioned automobile camera image grayscale reproduction test method.

[0075] Those skilled in the art will understand that all or part of the steps in the method for implementing the above-mentioned embodiments can be completed by instructing the relevant hardware through a program, and the program is stored in a storage medium, including several instructions for making a single-chip microcomputer, a chip or a processor (processor) perform all or part of the steps of the method described in each embodiment of the present invention. The aforementioned storage medium includes: U disk, mobile hard disk, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), disk or optical disk and other media that can store program codes.

[0076] The optional embodiments of the present invention are described in detail above in conjunction with the accompanying drawings. However, the embodiments of the present invention are not limited to the specific details in the above embodiments. Within the technical concept of the embodiments of the present invention, the technical scheme of the embodiments of the present invention can be subjected to a variety of simple modifications, and these simple modifications all belong to the protection scope of the embodiments of the present invention. It should also be noted that the various specific technical features described in the above specific embodiments can be combined in any suitable manner without contradiction. In order to avoid unnecessary repetition, the embodiments of the present invention will not further describe various possible combinations.

[0077] In addition, various embodiments of the present invention may be arbitrarily combined, and as long as they do not violate the concept of the embodiments of the present invention, they should also be regarded as the contents disclosed in the embodiments of the present invention.

Claims

1. A method for reproducing grayscale levels of automobile camera images, characterized in that: The method is performed by a self-checking end, which includes an experimental camera and a picture card shooting device, and the method includes: Recover the scene image of the current scene collected by the camera to be tested on the vehicle side; Analyze the scene image, and calibrate the parameter information of the camera to be tested based on the analysis result to obtain calibration parameters; The experimental camera parameters are adjusted in sequence based on the actual parameters and calibration parameters of the camera to be tested, and grayscale image card shooting is performed based on the adjusted experimental camera and image card shooting device; Grayscale brightness ratios are calculated based on the grayscale chart image corresponding to the actual parameters and the grayscale chart image corresponding to the calibration parameters, and grayscale level reproduction results of the actual parameters and grayscale level reproduction results of the calibration parameters are obtained based on the calculation results. The grayscale reproduction result of the camera to be tested is generated based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters.

2. The method according to claim 1, characterized in that The picture card shooting device comprises: The center gray back panel is used to place the grayscale chart card; An illumination cavity with the central grey back plate as the bottom surface of the cone; The lighting cavity is formed by a light-blocking baffle, and a fill light source is arranged on the light-blocking baffle; A shooting hole is provided at the vertex position of the cone formed by the lighting cavity and the central grey back plate, for setting an experimental camera.

3. The method according to claim 1, characterized in that The step of parsing the scene image and calibrating parameter information of the camera to be tested based on the parsing result to obtain calibration parameters includes: The horizontal field of view of the camera to be tested is calculated based on the scene image to obtain the calculated value of the horizontal field of view. The calculation rule is: Where, f is the focal length information of the scene image; w is the scene image width; H is the horizontal field of view information of the camera to be tested; Correcting the distortion parameters of the camera to be tested based on the calculated value of the horizontal field of view and the actual value of the horizontal field of view corresponding to the actual parameters of the camera to be tested; The parameter information corresponding to the camera to be tested after the distortion parameters are corrected is used as the calibration parameters.

4. The method according to claim 1, characterized in that: The method sequentially performs parameter adjustment of the experimental camera based on the actual parameters and calibration parameters of the camera to be tested, and performs grayscale picture card shooting based on the adjusted experimental camera and picture card shooting device, including: The experimental camera is adjusted in turn based on the actual parameters and calibration parameters of the camera to be tested. After completing one adjustment, grayscale card shooting is performed based on the experimental camera and the card shooting device to obtain the grayscale card shooting image corresponding to the actual parameters and the grayscale card shooting image corresponding to the calibration parameters.

5. The method according to claim 1, characterized in that: The grayscale brightness ratio is calculated based on the grayscale chart image corresponding to the actual parameters and the grayscale chart image corresponding to the calibration parameters, and the grayscale level reproduction result of the actual parameters and the grayscale level reproduction result under the calibration parameters are obtained based on the calculation results, including: In each captured image, each grayscale color block is identified based on an adaptive threshold segmentation algorithm; Identify the grayscale value of each grayscale color block, and assign a value to each grayscale color block based on the grayscale value; Determine whether the assignment results of each continuous grayscale color block are continuous values. If not, re-shoot the grayscale image card until the assignment results of the continuous grayscale color blocks are continuous values. Identify the brightness value of each grayscale color block, and calculate the brightness ratio of each grayscale color block based on the brightness value of each grayscale color block; The brightness difference between any two adjacent grayscale color blocks is calculated based on the brightness ratio of each grayscale color block, and a brightness difference set of actual parameters and a brightness difference set of calibration parameters are obtained respectively; Based on the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters, the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters are determined respectively.

6. The method according to claim 5, characterized in that The assigning of values ​​to each grayscale color block based on the grayscale value includes: Sort the grayscale color blocks based on their grayscale values. The higher the grayscale value, the higher the ranking. Based on the sorting results, each grayscale color block is assigned a value.

7. The method according to claim 5, characterized in that The identifying the brightness value of each grayscale color block and calculating the brightness ratio of each grayscale color block based on the brightness value of each grayscale color block includes: Based on the brightness value of each grayscale color block, calculate the brightness ratio between any two adjacent grayscale color blocks, traverse all grayscale color blocks, and obtain a brightness ratio set; Based on the brightness ratio set, convert and obtain a corresponding lightness ratio set; The method of calculating the brightness difference between any two adjacent grayscale color blocks based on the brightness ratio of each grayscale color block, and obtaining a brightness difference set of actual parameters and a brightness difference set of calibration parameters respectively, includes: Performing brightness assignment on the grayscale color block assigned a value of 1, and performing brightness assignment on each grayscale color block in turn based on the brightness assignment of the grayscale color block assigned a value of 1 and the brightness ratio set, to obtain brightness assignment results for each grayscale color block; Based on the brightness assignment results of each grayscale color block, the brightness difference between each two adjacent grayscale color blocks is calculated in turn to obtain a brightness difference set.

8. The method according to claim 5, characterized in that The step of respectively determining the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters based on the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters comprises: Traversing the brightness difference set of the actual parameters and the brightness difference set of the calibration parameters, respectively comparing each brightness difference with a preset difference threshold, and respectively recording the number of brightness difference values ​​corresponding to the actual parameters that are greater than the preset difference threshold and the number of brightness difference values ​​corresponding to the calibration parameters that are greater than the preset difference threshold, as a first brightness difference number and a second brightness difference number, respectively; If the first brightness difference value is greater than the preset difference value threshold, it means that the grayscale reproduction result of the actual parameter is qualified, otherwise, it is unqualified; If the second brightness difference value is greater than the preset difference value threshold, it means that the grayscale reproduction result under the calibration parameters is qualified, otherwise, it is unqualified; The grayscale reproduction result of the camera to be tested is generated based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters, including: If the grayscale reproduction result of the actual parameters is unqualified, but the grayscale reproduction result under the calibration parameters is qualified, it means that the test is invalid and the scene image is collected again to perform the test; If the grayscale reproduction result of the actual parameters is qualified, and the grayscale reproduction result under the calibration parameters is qualified, it means that the grayscale reproduction result of the camera to be tested is qualified; If the grayscale reproduction result of the actual parameters is unqualified, and the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified; If the grayscale reproduction result of the actual parameters is qualified, but the grayscale reproduction result under the calibration parameters is unqualified, it means that the grayscale reproduction result of the camera to be tested is unqualified, and the camera to be tested is adjusted based on the calibration parameters.

9. A car camera image grayscale reproduction test system, characterized in that: The system includes a vehicle terminal and a self-checking terminal that are communicatively connected to each other; The camera to be tested is arranged on the vehicle machine end, and the self-test end includes: Experimental camera and chart shooting device; A collection unit, used to recover the scene image of the current scene collected by the camera to be tested on the vehicle side; A calibration unit, used to analyze the scene image, and calibrate the parameter information of the camera to be tested based on the analysis result to obtain calibration parameters; The test unit is used to perform the experimental camera parameter adjustment in sequence based on the actual parameters and calibration parameters of the camera to be tested, and perform grayscale chart shooting based on the adjusted experimental camera and chart shooting device; A processing unit, used to calculate the grayscale brightness ratio based on the grayscale chart image corresponding to the actual parameters and the grayscale chart image corresponding to the calibration parameters, and obtain the grayscale level reproduction result of the actual parameters and the grayscale level reproduction result under the calibration parameters based on the calculation results; The determination unit is used to generate a grayscale reproduction result of the camera to be tested based on the grayscale reproduction result of the actual parameters and the grayscale reproduction result under the calibration parameters.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores instructions, which, when executed on a computer, enable the computer to execute the grayscale reproduction test method for an automobile camera image as described in any one of claims 1 to 8.

Images