Method and device for evaluating the placement position of a monocular camera

Abstract

The present application relates to a method and apparatus for evaluating the placement position of a monocular camera. The method includes: obtaining a captured image of the monocular camera, where the captured image includes a reference object disposed at a reference position of a vehicle; acquiring the image data of the captured image and the size data of the reference object, and determining a pixel accuracy value of the captured image based on the image data of the captured image and the size data of the reference object; and determining an evaluation score of the captured image according to a preset evaluation quantization formula and the pixel accuracy value. The present application provides a method and apparatus for evaluating the placement position of a monocular camera, which can improve the accuracy of evaluating the placement position of the monocular camera.

Description

Technical Field

[0001] This application relates to the field of intelligent transportation, and particularly to a method for evaluating the placement position of a monocular camera. Background Art

[0002] In the related art, in the field of intelligent transportation, several monocular cameras need to be installed on a vehicle to collect image information. However, when the placement position of the monocular camera is not accurate enough, error messages are easily generated. When evaluating the placement position of the monocular camera, a calibration board large enough needs to be selected to cover the camera projection plane, and there are also high requirements for the evaluation site. In this way, the calibration environment requirements for the monocular camera are high and rapid calibration of the monocular camera cannot be achieved.

[0003] Therefore, the position of the camera relative to the vehicle is crucial. How to effectively evaluate the placement position of the camera relative to the vehicle is what designers expect to achieve.

[0004] Application Content

[0005] To solve or partially solve the problems existing in the related art, this application provides a method and device for evaluating the placement position of a monocular camera, which can improve the accuracy of evaluating the placement position of the monocular camera.

[0006] The first aspect of this application provides a method for evaluating the placement position of a monocular camera, including:

[0007] Obtain a captured image of the monocular camera, where the captured image includes a reference object set at a reference position of the vehicle;

[0008] Obtain the image data of the captured image and the size data of the reference object, and determine the pixel accuracy value of the captured image based on the image data of the captured image and the size data of the reference object;

[0009] Determine the evaluation score of the captured image according to a preset evaluation quantization formula and the pixel accuracy value.

[0010] Optionally, determining the pixel accuracy value of the captured image based on the image data of the captured image and the actual data of the reference object includes:

[0011] Identify the feature edges of the reference object in the captured image from the image data, and determine the pixel length of the feature edges;

[0012] Determine the pixel accuracy value of the captured object according to the pixel length of the feature edges and the size data of the reference object.

[0013] Optionally, identifying the feature edges of the reference object in the captured image from the image data and determining the pixel length of the feature edges includes:

[0014] Identify the feature points of the reference object in the captured image, and determine the feature edges according to the feature points;

[0015] Based on the image data, obtain the two-dimensional coordinates corresponding to the feature points, and determine the pixel length corresponding to the feature edges according to the two-dimensional coordinates.

[0016] Optionally, after determining the evaluation score of the captured image, it includes:

[0017] If the evaluation score of the monocular camera is less than the first preset score threshold, output the first preset prompt, where the first preset prompt is used to indicate that the position of the monocular camera needs to be adjusted.

[0018] Optionally, after outputting the first preset prompt, it further includes:

[0019] If the evaluation score of the monocular camera is less than the second preset score threshold, output the second preset prompt, where the second preset prompt is used to indicate that the monocular camera needs to be adjusted significantly;

[0020] If the evaluation score of the monocular camera is greater than the second preset score threshold and less than the first preset score threshold, output the third preset prompt, where the third preset prompt is used to indicate that the monocular camera needs fine-tuning.

[0021] Optionally, the method includes:

[0022] Before adjusting the monocular camera, record the historical evaluation score of the monocular camera;

[0023] After adjusting the monocular camera, obtain the current evaluation score of the monocular camera;

[0024] According to the current evaluation score and the historical evaluation score, determine the predicted adjustment angle of the monocular camera, where the predicted adjustment angle is used to provide a reference adjustment angle for the monocular camera.

[0025] Optionally, if the evaluation score of the monocular camera does not meet the preset score threshold, after the evaluation result of the placement position of the monocular camera is determined to need adjustment, it includes:

[0026] Adjust the placement position of the monocular camera so that the evaluation score of the monocular camera is greater than the preset score threshold to complete the calibration of the monocular camera.

[0027] Optionally, the vehicle reference position includes at least one of the front end of the vehicle, the rear end of the vehicle, the front side of the vehicle, and the lane line position.

[0028] The second aspect of the present application provides an evaluation device for the placement position of a monocular camera, including:

[0029] A captured image reading unit, configured to obtain the captured image of the monocular camera, where the captured image includes a reference object arranged at the vehicle reference position;

[0030] A processing unit, configured to obtain the image data of the captured image and the dimension data of the reference object, and determine the pixel accuracy value of the captured image based on the image data of the captured image and the dimension data of the reference object;

[0031] An evaluation unit, configured to determine the evaluation score of the captured image according to a preset evaluation quantization formula and the pixel accuracy value.

[0032] A third aspect of the present application provides an electronic device, including:

[0033] A processor; and

[0034] A memory, storing executable code thereon, which when executed by the processor, causes the processor to execute the method as described above.

[0035] A fourth aspect of the present application provides a computer-readable storage medium, storing executable code thereon, which when executed by the processor of the electronic device, causes the processor to execute the method as described above.

[0036] The technical solution provided by the present application may include the following beneficial effects: In the first aspect, the present application replaces the large calibration ruler used in the traditional camera calibration method by setting a reference object in a certain road scene. According to the captured image of a certain road scene by a monocular camera, the image data of the captured image and the dimension data of the reference object are obtained, and the pixel accuracy value of the captured image is determined. According to the preset evaluation quantization formula and the pixel accuracy value, the evaluation score of the monocular camera is determined. The user can adjust the placement position of the monocular camera according to the evaluation score output by the reference system under any scene conditions, thus getting rid of the bondage of the calibration board and the site in the existing monocular camera calibration process, realizing the solution for quickly analyzing the relative position between the monocular camera and the vehicle, and improving the efficiency and accuracy of evaluating the placement position of the monocular camera.

[0037] In the second aspect, after determining the evaluation score of the monocular camera, relevant prompts for camera adjustment can also be output according to the evaluation score. If the evaluation score of the monocular camera is less than the first preset score threshold, the first preset prompt is output, and the first preset prompt is used to indicate that the position of the monocular camera needs to be adjusted. If the evaluation score of the monocular camera is less than the second preset score threshold, the second preset prompt is output, and the second preset prompt is used to indicate that the monocular camera needs to be adjusted significantly; if the evaluation score of the monocular camera is greater than the second preset score threshold and less than the first preset score threshold, the third preset prompt is output, and the third preset prompt is used to indicate that the monocular camera needs fine-tuning. The user can change the camera adjustment amplitude according to the preset prompt, thereby improving the efficiency of monocular camera calibration.

[0038] It should be understood that the above general description and the following detailed description are only exemplary and explanatory, and cannot limit the present application. Description of the Drawings

[0039] The above and other objects, features, and advantages of the present application will become more apparent by describing the exemplary embodiments of the present application in more detail with reference to the accompanying drawings. In the exemplary embodiments of the present application, the same reference numerals generally represent the same components.

[0040] Figure 1 It is a schematic diagram of the application environment of the method for evaluating the placement position of a monocular camera shown in the embodiments of the present application;

[0041] Figure 2 It is a schematic flowchart of the method for evaluating the placement position of a monocular camera shown in the embodiments of the present application;

[0042] Figure 3 It is a schematic structural diagram of the device for evaluating the placement position of a monocular camera shown in the embodiments of the present application;

[0043] Figure 4 It is a schematic structural diagram of an electronic device shown in the embodiments of the present application. Detailed Embodiments

[0044] The embodiments of the present application will be described in more detail below with reference to the accompanying drawings. Although the embodiments of the present application are shown in the drawings, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments set forth herein. On the contrary, these embodiments are provided to make the present application more thorough and complete, and to fully convey the scope of the present application to those skilled in the art.

[0045] The terms used in the present application are for the purpose of describing specific embodiments only and are not intended to limit the present application. The singular forms "a", "the", and "said" used in the present application and the appended claims are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" as used herein refers to and includes any and all possible combinations of one or more of the associated listed items.

[0046] It should be understood that although the terms "first", "second", "third", etc. may be used in the present application to describe various information, such information should not be limited to these terms. These terms are only used to distinguish the same type of information from each other. For example, without departing from the scope of the present application, the first information may also be referred to as the second information, and similarly, the second information may also be referred to as the first information. Thus, the features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the present application, "a plurality" means two or more unless otherwise specifically defined.

[0047] In view of the above problems, an embodiment of the present application provides a method and device for evaluating the placement position of a monocular camera, which can improve the accuracy of the method for evaluating the placement position of the monocular camera.

[0048] The technical solutions of the embodiments of the present application will be described in detail below with reference to the accompanying drawings.

[0049] The positioning method of the autonomous vehicle provided by the present application can be applied to an application environment as Figure 1 shown. Among them, the terminal 102 communicates with the server 104 through a network. The terminal 102 is any device with computing hardware that can support and execute a software product corresponding to a game, and can be, but is not limited to, various personal computers, laptop computers, smart phones, tablet computers, in-vehicle computers, and portable wearable devices. The terminal 102 has a shooting interface for displaying the monocular camera (i.e., the real-time image captured by the monocular camera) and an evaluation score for the placement position of the monocular camera, where the evaluation score is displayed in real time. The server 104 is used to receive relevant data of the monocular camera.

[0050] In the present application, by obtaining the pixel accuracy value of a reference object, which is set at a preset vehicle reference position; according to a preset evaluation quantization formula and the pixel accuracy value of the reference object, an evaluation score of the monocular camera is obtained; based on a preset score threshold and the evaluation score of the monocular camera, the placement position of the monocular camera is evaluated. The solution provided by the present application can reasonably evaluate the placement position of the monocular camera according to the pixel accuracy value of the reference object.

[0051] In one embodiment, the server 104 is connected to a monocular camera, which is set at a suitable position of the vehicle. For example, the monocular camera can be set at the front windshield of the vehicle or at the rearview mirror of the vehicle, so that the monocular camera can capture a shooting image including a road sign during the automatic driving process of the vehicle. However, the imaging effect of the monocular camera is also related to the position of the monocular camera. When the placement position of the monocular camera is not accurate enough, error messages are easily generated. When evaluating the placement position of the monocular camera, a calibration board large enough to cover the camera projection plane needs to be selected, and there are also high requirements for the evaluation site. In this way, the calibration environment requirements for the monocular camera are high and rapid calibration of the monocular camera cannot be achieved. By installing the camera in different ways, different external parameters can be obtained. The evaluation score of the current installation method can be obtained through data such as the size and accuracy of the effective range. By evaluating the shooting image of the monocular camera, the external parameters corresponding to different positions are converted into the evaluation score of the monocular camera, and the evaluation score provides a reference for the installer.

[0052] Figure 2 It is a schematic flowchart of the method for evaluating the placement position of the monocular camera shown in the embodiment of the present application.

[0053] See Figure 2, A method for evaluating the placement position of a monocular camera, including:

[0054] Step S201: Obtain a captured image of the monocular camera, where the captured image includes a reference object disposed at a reference position of the vehicle.

[0055] Capture a captured image of a certain road scene. The captured image can be a series of captured images within a certain time period or a certain captured image with a high scoring quality. In step S201, by placing a reference object at the reference position of the vehicle, it can replace the traditional camera calibration ruler and camera calibration scene.

[0056] In step S201, the reference object can be a calibration object such as a triangular ruler that can know the actual length of the reference object. Since the monocular camera is mainly used to capture map elements around the autonomous vehicle. The preset vehicle reference position is fixed around the vehicle.

[0057] In one embodiment, the vehicle reference position is the placement position of the reference object relative to the vehicle. The vehicle reference position includes at least one of the vehicle front end, the vehicle rear end, the vehicle side front, and the lane line position.

[0058] Step S202: Obtain the image data of the captured image and the size data of the reference object, and determine the pixel accuracy value of the captured image based on the image data of the captured image and the size data of the reference object.

[0059] Accuracy is generally used to represent the relationship between the measured value and the actual value, and the pixel accuracy value is generally used to represent the situation of the number of pixels required to display the target reference object. When the reference object is closer to the monocular camera (i.e., at the lower part of the captured image of the monocular camera), the reference object will occupy more pixels in the captured image of the monocular camera (i.e., for the same captured image, the higher the pixel accuracy value, the shorter the distance of each pixel), while when the reference object is far from the monocular camera, the reference object will occupy fewer pixels in the captured image of the monocular camera (the distance of each pixel becomes larger). Therefore, for monocular cameras with different placement positions, the pixel accuracy values of the same reference object at the same position are different, and the better the placement position, the higher the pixel accuracy value. According to the matrix expression form of the camera extrinsic parameters, the extrinsic parameters correspond one-to-one with the pixel accuracy value.

[0060] Specifically, the coordinates of a pixel in the world coordinate system and the coordinates of the pixel in the pixel coordinate system can be obtained. Based on the principle of monocular camera photography, a conversion formula regarding the camera internal parameters, camera external parameters, world coordinate system, and pixel coordinates can be established. And according to the pixel coordinates and the actual coordinates, the pixel accuracy value can be obtained, and the internal parameters are also related to the factory coefficients of the camera. Therefore, the corresponding relationship between the pixel accuracy value and the monocular camera external parameters can be obtained, and the monocular camera external parameters are related to the placement position of the monocular camera, so that the evaluation score can be calculated for the pixel accuracy value of the captured image, and thus an evaluation method for the placement position of the monocular camera can be established.

[0061] In one embodiment, in step S202, obtaining the pixel accuracy value of the reference object includes: identifying the feature edges of the reference object in the captured image from the image data, and determining the pixel length of the feature edges; according to the pixel length of the feature edges and the size data of the reference object, determining the pixel accuracy value of the captured object.

[0062] In this embodiment, the feature edge can be the complete edge of the reference object with known size data. If the reference object is a triangular ruler, the feature edge is the edge of the triangular ruler with known size data. For example, if the length of the hypotenuse of the triangular ruler is known, the hypotenuse length of the triangular ruler can be used.

[0063] Specifically, identifying the feature edges of the reference object in the captured image from the image data and determining the pixel length of the feature edges includes: identifying the feature points of the reference object in the captured image, and determining the feature edges according to the feature points; obtaining the two-dimensional coordinates corresponding to the feature points based on the image data, and determining the pixel length corresponding to the feature edges according to the two-dimensional coordinates.

[0064] In this embodiment, to identify the feature points of the reference object, the point cloud algorithm can be used to model the captured image of the reference object, so as to identify the pixel feature points in the reference captured image and obtain the two-dimensional coordinates of the pixel feature points; the grid algorithm can also be used to identify the feature points, and then a two-dimensional coordinate system is constructed according to the captured image to obtain the coordinates of the feature points.

[0065] In one embodiment, after obtaining the feature points, the two-dimensional coordinates of the feature points may not be obtained, and the pixel length of the feature edges can be directly obtained according to the feature points on the captured image.

[0066] Specifically, the preset evaluation quantization formula is obtained as formula (1):

[0067] P1 = D pixel / D real (1);

[0068] Where D pixel is used to represent the pixel length of the reference object, D real is used to represent the actual length of the reference object, and P1 is used to represent the pixel accuracy value.

[0069] In one embodiment, the reference object is a triangular ruler. Obtaining the pixel accuracy value of the reference object includes identifying the vertices of the triangular ruler, obtaining the lengths of the three sides of the triangular ruler, and obtaining the pixel lengths of the three sides in the captured image. Since the lengths of the sides of the triangular ruler are known, the actual lengths of the sides of the triangular ruler can be obtained, and according to the ratio of the actual length to the pixel length of the triangular ruler, the pixel accuracy values corresponding to the three sides are obtained.

[0070] In one embodiment, the reference object can be a map element with known dimension data such as a lane line. For example, the distance between adjacent lane lines is fixed and unique, and the distance between adjacent lane lines can be selected as the dimension data of the lane line; obtain the pixel width of the distance between the lane lines in the captured image; according to the ratio of the dimension data of the lane line distance to the pixel length, the pixel accuracy value of the reference object can be obtained.

[0071] Step S203: Determine the evaluation score of the captured image according to the preset evaluation quantization formula and the pixel accuracy value.

[0072] The preset evaluation quantization formula is used to convert the pixel accuracy value into a visual measurement score (i.e., the evaluation score in this application). The higher the pixel accuracy value, the better the evaluation score of the camera setting position. According to the monocular camera to adjust the camera position, the pixel accuracy value of the reference object can be calculated in real time, and the evaluation score can be calculated according to the pixel accuracy value. The evaluation score can be output in real time according to the shooting position of the monocular camera.

[0073] In one embodiment, step S203 includes: obtaining the pixel accuracy value of the feature edge, obtaining the evaluation score of the reference object according to the feature edge and the preset evaluation quantization formula; obtaining the average value of the evaluation scores between the reference objects according to the evaluation score of the reference object and the number of reference objects, and the average value of the evaluation scores is the evaluation score of the monocular camera.

[0074] Since the pixel accuracy value can reflect the display accuracy of the reference object in the captured image, the preset evaluation quantization formula for evaluating the pixel accuracy value is formula (2). Formula (2) includes:

[0075]

[0076] Where p and pi represent the pixel accuracy values of each side, and i represents the i-th side. According to formula (2), the score corresponding to each side length can be calculated. According to the evaluation scores corresponding to each side, the average score of the feature edges of the reference object is obtained, and the average score is used as the evaluation score of the camera, or the average value of the evaluation scores of several reference objects is used as the evaluation score of the camera. The range of this score is from 0 to 100 points.

[0077] In one embodiment, the preset evaluation quantization formula may also adopt an evaluation function. The evaluation function is used to output a predicted score according to the pixel accuracy value, and score the pixel accuracy value according to the range of the evaluation score output result.

[0078] In one embodiment, the monocular camera is divided into multiple regions according to the preset vehicle reference position. Each region corresponds to an evaluation score, and the evaluation score is displayed in real time in the region. Specifically, step S202 includes: within the region, obtaining the pixel accuracy value of the feature edge, and according to the feature edge and the preset evaluation quantization formula, obtaining the evaluation score of the reference object in each region; according to the evaluation score of each region, presenting the shooting effect of each region.

[0079] In one embodiment, after determining the evaluation score of the captured image, it includes: if the evaluation score of the monocular camera is less than the first preset score threshold, outputting a first preset prompt, where the first preset prompt is used to indicate that the position of the monocular camera needs to be adjusted.

[0080] In this embodiment, the first preset score threshold is obtained according to the captured image of the calibrated monocular camera. The first preset prompt may be the display text on the screen or a piece of prompt sound.

[0081] In this embodiment, if the evaluation score of the monocular camera is less than the first preset score threshold, the result of the first preset prompt is that the placement position of the monocular camera needs to be adjusted; if the evaluation score of the monocular camera is greater than the first preset score threshold, the evaluation result of the position of the monocular camera is that the placement position of the monocular camera is correct.

[0082] In one embodiment, after outputting the first preset prompt, it further includes: if the evaluation score of the monocular camera is less than the second preset score threshold, outputting a second preset prompt, where the second preset prompt is used to indicate that the monocular camera needs to be adjusted significantly; if the evaluation score of the monocular camera is greater than the second preset score threshold and less than the first preset score threshold, outputting a third preset prompt, where the third preset prompt is used to indicate that the monocular camera needs to be fine-tuned.

[0083] Specifically, the user can try to adjust the monocular camera angle from multiple angles according to the third preset prompt and the second preset prompt, so that the evaluation score of the camera is greater than the first preset score threshold. If the user receives the second preset prompt, according to the second preset prompt, the user greatly adjusts the monocular camera, traverses the first preset adjustment angle set (the first preset adjustment angle set includes several first preset angles for greatly adjusting the monocular camera), so that the monocular camera displays the third preset prompt; according to the third preset prompt, the user slightly adjusts the position of the monocular camera, traverses the second preset adjustment angle set (the second preset adjustment angle set includes several second preset angles for slightly adjusting the monocular camera), so that the evaluation score of the monocular camera is greater than the first preset threshold. At this time, the monocular camera does not display the first preset prompt, the second preset prompt and the third preset prompt.

[0084] In one embodiment, the method includes: before adjusting the monocular camera, recording the historical evaluation score of the monocular camera; after adjusting the monocular camera, obtaining the current evaluation score of the monocular camera; according to the current evaluation score and the historical evaluation score, determining the predicted adjustment angle of the monocular camera, and the predicted adjustment angle is used to provide a reference adjustment angle for the monocular camera.

[0085] Specifically, the historical evaluation score of the monocular camera corresponds one-to-one with the historical adjustment angle of the monocular camera, and the difference between the current evaluation score and the historical evaluation score of the monocular camera is obtained; in the order of the difference from small to large, the angle with the largest difference from the current evaluation score of the monocular camera is selected as the predicted angle.

[0086] In one embodiment, the captured image can be divided into four regions, and each region corresponds to a reference object. Then the evaluation score of the captured image includes the evaluation scores of the four regions. Adjusting the angle of the monocular camera according to the difference between the evaluation score and the preset score threshold includes: respectively obtaining the differences between the four evaluation scores and the first preset score threshold, and moving the angle of the monocular camera in the direction of the region with the larger difference. During the movement, the four regions are displayed on the camera screen in real time, and the camera movement is adjusted according to the differences of the four regions until the evaluation scores of the four regions of the camera reach the first preset score threshold.

[0087] In one embodiment, the evaluation score includes the evaluation scores of multiple regions. An axis can be established according to the preset reference position where the reference object is located, the reference direction where the reference object is located is displayed according to the axis, and the evaluation score is displayed on the axis. The position of the monocular camera can be adjusted according to the evaluation score.

[0088] In one embodiment, a preset area divides the shooting area of the monocular camera into an upper area and a lower area. Each area corresponds to a preset reference position and a reference object is set at this position. The coordinate axis is set on the boundary line between the upper area and the lower area, and evaluation scores are displayed at both ends of the boundary line. If the score of the upper area is less than the lower area, the monocular camera is moved upward; if the score of the lower area is less than the lower area, the monocular camera is moved downward.

[0089] In one embodiment, the preset area divides the shooting area of the monocular camera into more than two areas. Each area corresponds to a corresponding preset reference area, and a reference object is set in each preset reference area. Each area corresponds to a corresponding evaluation score according to the reference object in this area. When the evaluation score is less than the preset threshold, the camera is adjusted from the direction of this area.

[0090] In one embodiment, when the position of the monocular camera is adjusted, the evaluation score of the monocular camera is displayed in real time according to the adjusted position of the monocular camera. By displaying the evaluation score of the monocular camera in real time, it can assist the staff to adjust the camera angle.

[0091] The technical solution provided by this application may include the following beneficial effects: In the first aspect, this application replaces the large calibration ruler used in the traditional camera calibration method by setting reference objects in a certain road scene. According to the captured image of a certain road scene by the monocular camera, the image data of the captured image and the size data of the reference object are obtained, and the pixel accuracy value of the captured image is determined. According to the preset evaluation quantization formula and the pixel accuracy value, the evaluation score of the monocular camera is determined. Users can adjust the placement position of the monocular camera according to the evaluation score output by the reference system under any scene conditions, thus getting rid of the constraints on the calibration board and the site in the existing monocular camera calibration process, realizing a solution for quickly analyzing the relative position between the monocular camera and the vehicle, and improving the efficiency and accuracy of evaluating the placement position of the monocular camera.

[0092] In the second aspect, after determining the evaluation score of the monocular camera, relevant prompts for camera adjustment can also be output according to the evaluation score. If the evaluation score of the monocular camera is less than the first preset score threshold, the first preset prompt is output, and the first preset prompt is used to indicate that the position of the monocular camera needs to be adjusted. If the evaluation score of the monocular camera is less than the second preset score threshold, the second preset prompt is output, and the second preset prompt is used to indicate that the monocular camera needs to be adjusted significantly; if the evaluation score of the monocular camera is greater than the second preset score threshold and less than the first preset score threshold, the third preset prompt is output, and the third preset prompt is used to indicate that the monocular camera needs fine-tuning. Users can change the camera adjustment amplitude according to the preset prompts, thereby improving the efficiency of monocular camera calibration.

[0093] Figure 3It is a schematic structural diagram of an evaluation device for the placement position of a monocular camera shown in an embodiment of the present application.

[0094] See Figure 3 , Figure 3 It includes a captured image reading unit 301, a processing unit 302, and an evaluation unit 303, where:

[0095] The captured image reading unit 301 is used to obtain the captured image of the monocular camera, and the captured image includes a reference object set at the vehicle reference position.

[0096] The captured image of a certain road scene is collected, which can be a series of captured images within a certain time period, or a certain captured image with a higher scoring quality of the captured image. The acquisition unit 301 replaces the function of the calibration board in the original monocular camera by placing a reference object at the preset vehicle reference position, calculates the pixel accuracy value of the reference object, and adjusts the placement position of the monocular camera according to the real-time pixel accuracy value. The reference object can be a calibration object such as a triangular ruler that can know the actual length of the reference object. Since the monocular camera is mainly used to capture map elements around the autonomous vehicle. The preset vehicle reference position is fixed around the vehicle. The vehicle reference position is the placement position of the reference object relative to the vehicle. By placing a reference object at the vehicle reference position, the traditional camera calibration ruler and camera calibration scene can be replaced. The vehicle reference position includes at least one of the vehicle front end, the vehicle rear end, the vehicle side front, and the lane line position.

[0097] The second unit 302 is used to obtain the image data of the captured image and the size data of the reference object, and determine the pixel accuracy value of the captured image based on the image data of the captured image and the size data of the reference object.

[0098] Accuracy is generally used to represent the relationship between the measured value and the actual value, and the pixel accuracy value is generally used to represent the situation of the number of pixels required to display the target reference object. At the position where the reference object is closer to the monocular camera (i.e., below the captured image of the monocular camera), the reference object will occupy more pixels in the captured image of the monocular camera (that is, for the same captured image, the higher the pixel accuracy value, the shorter the distance of each pixel), while at the position where the reference object is far from the monocular camera, the reference object will occupy fewer pixels in the captured image of the monocular camera (the distance of each pixel becomes larger). Therefore, for monocular cameras with different placement positions, the pixel accuracy values of the same position and the same reference object are different, and the better the placement position, the higher the pixel accuracy value. The second unit 302 replaces the function of the calibration board in the original monocular camera by placing a reference object at the preset vehicle reference position, calculates the pixel accuracy value of the reference object, and adjusts the placement position of the monocular camera according to the real-time pixel accuracy value.

[0099] The reference object can be a calibration object such as a triangular ruler that can know the actual length of the reference object. Since the monocular camera is mainly used to capture map elements around the autonomous vehicle. The preset vehicle reference position is fixed around the vehicle.

[0100] In one embodiment, the vehicle reference position mainly uses the position where the monocular camera obtains the captured image. The vehicle reference position includes at least one of the vehicle front end, the vehicle rear end, the vehicle side front, and the lane line position.

[0101] In one embodiment, obtaining the pixel accuracy value of the reference object includes: identifying the key points of the reference object, and obtaining the characteristic edges of the reference object according to the key points of the reference object; obtaining the actual length of the characteristic edge and the pixel length of the characteristic edge, and obtaining the pixel accuracy value of the reference object according to the actual length and the pixel length of the characteristic edge.

[0102] Specifically, the formula for obtaining the pixel accuracy value is formula (1):

[0103] P1 = D pixel / D real ; (1);

[0104] Wherein, D pixel is used to represent the pixel length of the reference object, D real is used to represent the actual length of the reference object, and P1 is used to represent the pixel accuracy value.

[0105] In one embodiment, the reference object is a triangular ruler. Obtaining the pixel accuracy value of the reference object includes identifying the vertices of the triangular ruler, obtaining the complete sides of the three sides in the triangular ruler, and obtaining the pixel lengths of the three complete sides in the captured image. Since the lengths of the complete sides of the triangular ruler are known, the actual side lengths of the triangular ruler can be obtained, and according to the ratio of the actual length and the pixel length of the complete side, the pixel accuracy value corresponding to the complete side is obtained, and the average pixel accuracy value of the complete side is obtained to obtain the pixel accuracy value of the captured image.

[0106] In one embodiment, the reference object can be a map element with known size data such as a lane line. For example, the distance between adjacent lane lines is fixed and unique, and the distance between adjacent lane lines can be selected as the size data of the lane line; obtaining the pixel width of the distance between lane lines in the captured image; according to the ratio of the size data of the lane line distance and the pixel length, the pixel accuracy value of the reference object can be obtained.

[0107] Among them, the preset evaluation quantization formula is used to convert the pixel accuracy value into a visual measurement score (i.e., the evaluation score in this application). The higher the pixel accuracy value, the better the quality of the camera setting position, and the higher the corresponding evaluation score. According to the monocular camera to adjust the camera position, the pixel accuracy value of the reference object can be calculated in real time, and the evaluation score can be obtained according to the pixel accuracy value, and the evaluation score is displayed in real time according to the camera position.

[0108] The evaluation unit 303 determines the evaluation score of the captured image according to the preset evaluation quantization formula and the pixel accuracy value.

[0109] Obtain the pixel accuracy value of the feature edge, obtain the evaluation score of the reference object according to the feature edge and the preset evaluation quantization formula; obtain the average value of the evaluation scores between the reference objects according to the evaluation score of the reference object and the number of reference objects, and the average value of the evaluation scores is the evaluation score of the monocular camera.

[0110] Since the pixel accuracy value can reflect the display accuracy of the reference object in the captured image, the preset evaluation quantization formula for evaluating the pixel accuracy value is formula (2), and formula (2) includes:

[0111]

[0112] Among them, p and pi represent the pixel accuracy values of each edge, and i represents the i-th edge. According to formula (2), the score corresponding to each edge length can be calculated, and the average score of the feature edges of the reference object can be obtained according to the evaluation score corresponding to each edge, and the average score is used as the evaluation score of the camera, or the average value of the evaluation scores of several reference objects is used as the evaluation score of the camera, and the range of this score is from 0 to 100 points.

[0113] In one embodiment, the preset evaluation quantization formula can also adopt an evaluation function, and the evaluation function is used to output a predicted score according to the pixel accuracy value, and score the pixel accuracy value according to the range of the evaluation score output result.

[0114] In one embodiment, the monocular camera is divided into multiple regions according to the preset vehicle reference position. Each region corresponds to an evaluation score. The evaluation score is displayed in real time in this region. Specifically, it includes: within the evaluation region, obtain the pixel accuracy value of the feature edge, and obtain the evaluation score of the reference object in each evaluation region according to the feature edge and the preset evaluation quantization formula; according to the evaluation score of each evaluation region.

[0115] In one embodiment, the device further includes: a display unit, and the display unit is used to output a first preset prompt and a second preset prompt according to the evaluation score.

[0116] In one embodiment, after determining the evaluation score of the captured image, it includes: if the evaluation score of the monocular camera is less than the first preset score threshold, output a first preset prompt, where the first preset prompt is used to indicate that the position of the monocular camera needs to be adjusted.

[0117] In this embodiment, the first preset score threshold is obtained based on the captured image of the calibrated monocular camera. The first preset prompt can be the display text on the screen or a prompt tone.

[0118] In this embodiment, if the evaluation score of the monocular camera does not meet the preset score threshold, the evaluation result of the placement position of the monocular camera is: the placement position of the monocular camera needs to be adjusted; if the evaluation score of the monocular camera is greater than the preset score threshold, the evaluation result of the position of the monocular camera is: the placement position of the monocular camera is correct.

[0119] In one embodiment, after outputting the first preset prompt, it further includes: if the evaluation score of the monocular camera is less than the second preset score threshold, output a second preset prompt, where the second preset prompt is used to indicate that the monocular camera needs to be adjusted significantly; if the evaluation score of the monocular camera is greater than the second preset score threshold and less than the first preset score threshold, output a third preset prompt, where the third preset prompt is used to indicate that the monocular camera needs fine-tuning.

[0120] Specifically, the preset score threshold includes the first preset score threshold and the second preset score threshold. When the evaluation score of the monocular camera is less than the first preset score threshold, the camera screen displays that it needs to be reinstalled. When the evaluation score of the monocular camera is between the first preset score threshold and the second preset score threshold, adjust the camera angle so that its evaluation score is within the preset range.

[0121] In one embodiment, the method includes: before adjusting the monocular camera, record the historical evaluation score of the monocular camera; after adjusting the monocular camera, obtain the current evaluation score of the monocular camera; based on the current evaluation score and the historical evaluation score, determine the predicted adjustment angle of the monocular camera, where the predicted adjustment angle is used to provide a reference adjustment angle for the monocular camera.

[0122] Specifically, the position of the monocular camera can be adjusted by trying multiple angles, so that the evaluation score of the monocular camera can be adjusted from the original score to the preset score threshold.

[0123] In one embodiment, when obtaining the evaluation score of the reference object, it further includes: obtaining the difference between the evaluation score of the reference object and the preset score threshold; arranging the monocular cameras in descending order of the difference and starting to adjust from the vehicle reference position corresponding to the reference object.

[0124] Specifically, the evaluation score includes the scores of four reference positions where the reference object is located. Adjusting the angle of the monocular camera according to the difference between the evaluation score and the preset score threshold includes obtaining the difference between the evaluation score and the preset score threshold, and moving the angle of the monocular camera in the direction with the larger difference. During the movement, the camera screen displays four regions in real time, and adjusts the camera movement according to the differences in the four regions until the score of the camera reaches the passing score.

[0125] In one embodiment, the evaluation score includes the evaluation scores of multiple evaluation regions. Coordinate axes can be established according to the preset reference positions where the reference object is located, and the reference direction of the reference object can be displayed according to the coordinate axes. The evaluation scores are displayed on the coordinate axes, and the position of the monocular camera can be adjusted according to the evaluation scores.

[0126] Specifically, according to different preset regions, coordinate axes are established and the coordinate axes are displayed on the regional demarcation lines.

[0127] In one embodiment, the preset region divides the shooting region of the monocular camera into two parts, namely the upper region and the lower region. Each region corresponds to a preset reference position and a reference object is set at this position. The coordinate axes are set on the demarcation line between the upper region and the lower region, and the evaluation scores are displayed at both ends of the demarcation line. If the score of the upper region is less than the lower region, the monocular camera is moved upward; if the score of the lower region is less than the lower region, the monocular camera is moved downward.

[0128] In one embodiment, the preset region divides the shooting region of the monocular camera into more than two regions. Each region corresponds to a corresponding preset reference area, and a reference object is set in each preset reference area. Each area corresponds to a corresponding evaluation score according to the reference object in this area. In the case where the evaluation score is less than the preset threshold, the camera is adjusted from the direction of this region.

[0129] In one embodiment, when the position of the monocular camera is adjusted, the evaluation score of the monocular camera is displayed in real time according to the adjusted position of the monocular camera. By displaying the evaluation score of the monocular camera in real time, it can assist the staff in adjusting the camera angle.

[0130] Regarding the device in the above embodiments, the specific manner in which each module performs operations has been described in detail in the embodiments related to the method, and will not be elaborated here in detail.

[0131] Figure 4 It is a schematic structural diagram of an electronic device shown in an embodiment of the present application.

[0132] See Figure 4 Electronic device 400 includes a memory 410 and a processor 420.

[0133] The processor 420 can be a Central Processing Unit (CPU), or can also be other general-purpose processors, Digital Signal Processors (DSPs), Application Specific Integrated Circuits (ASICs), Field-Programmable Gate Arrays (FPGAs), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general-purpose processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0134] The memory 410 can include various types of storage units, such as system memory, read-only memory (ROM), and permanent storage devices. Among them, the ROM can store static data or instructions required by the processor 420 or other modules of the computer. The permanent storage device can be a read-write storage device. The permanent storage device can be a non-volatile storage device that does not lose the stored instructions and data even when the computer is powered off. In some embodiments, the permanent storage device uses a mass storage device (such as a magnetic or optical disk, flash memory) as the permanent storage device. In some other embodiments, the permanent storage device can be a removable storage device (such as a floppy disk, optical drive). The system memory can be a read-write storage device or a volatile read-write storage device, such as dynamic random access memory. The system memory can store some or all of the instructions and data required by the processor during operation. In addition, the memory 410 can include any combination of computer-readable storage media, including various types of semiconductor storage chips (such as DRAM, SRAM, SDRAM, flash memory, programmable read-only memory), and magnetic disks and / or optical disks can also be used. In some embodiments, the memory 410 can include a removable storage device that is readable and / or writable, such as a compact disc (CD), read-only digital versatile disc (such as DVD-ROM, dual-layer DVD-ROM), read-only Blu-ray disc, super density disc, flash memory card (such as SD card, min SD card, Micro-SD card, etc.), magnetic floppy disk, etc. The computer-readable storage medium does not include carrier waves and instantaneous electronic signals transmitted wirelessly or by wire.

[0135] An executable code is stored on the memory 410, and when the executable code is processed by the processor 420, it can cause the processor 420 to execute some or all of the methods described above.

[0136] In addition, the method according to the present application can also be implemented as a computer program or a computer program product, which includes computer program code instructions for performing some or all of the steps in the above method of the present application.

[0137] Alternatively, the present application can also be implemented as a computer-readable storage medium (or a non-transitory machine-readable storage medium or a machine-readable storage medium), on which executable code (or a computer program or computer instruction code) is stored. When the executable code (or the computer program or computer instruction code) is executed by a processor of an electronic device (or a server, etc.), the processor is caused to execute some or all of the steps of the above method according to the present application.

[0138] The embodiments of the present application have been described above. The above description is exemplary and not exhaustive, and is also not limited to the disclosed embodiments. Many modifications and variations are obvious to those of ordinary skill in the art in the technical field without departing from the scope and spirit of the described embodiments. The selection of the terms used herein is intended to best explain the principles of the embodiments, the practical application or the improvement of the technology in the market, or to enable other ordinary skill in the art in the technical field to understand the embodiments disclosed herein.

Claims

1. An evaluation method for the placement position of a monocular camera, characterized in that, Including: Obtain a captured image of a single monocular camera, where the captured image includes a reference object set at a vehicle reference position; Obtain the image data of the captured image of the single monocular camera and the size data of the reference object, and based on the image data of the captured image and the size data of the reference object, determine the pixel accuracy value of the captured image; It includes: obtaining the feature edges of the reference object in the captured image according to the image data, and determining the pixel length of the feature edges; determining the pixel accuracy value of the reference object according to the pixel length of the feature edges and the size data of the reference object; Determine the evaluation score of the captured image according to a preset evaluation quantization formula and the pixel accuracy value; including, obtaining the pixel accuracy value of the feature edges, obtaining the evaluation score of the reference object according to the feature edges and the preset evaluation quantization formula; obtaining the average value of the evaluation scores between the reference objects according to the evaluation score of the reference object and the number of reference objects, and the average value of the evaluation scores is the evaluation score of the monocular camera; Output a corresponding adjustment amplitude prompt according to the evaluation score of the captured image of the single monocular camera.

2. The method according to claim 1, wherein The obtaining the feature edges of the reference object in the captured image according to the image data, and determining the pixel length of the feature edges includes: Identifying the feature points of the reference object in the captured image, and determining the feature edges according to the feature points; Obtaining the two-dimensional coordinates corresponding to the feature points based on the image data, and determining the pixel length corresponding to the feature edges according to the two-dimensional coordinates.

3. The method according to claim 1, characterized in that, After determining the evaluation score of the captured image, it includes: If the evaluation score of the monocular camera is less than the first preset score threshold, output a first preset prompt, where the first preset prompt is used to indicate that the position of the monocular camera needs to be adjusted.

4. The method according to claim 3, characterized in that, After outputting the first preset prompt, it further includes: If the evaluation score of the monocular camera is less than the second preset score threshold, output a second preset prompt, where the second preset prompt is used to indicate that the monocular camera needs to be adjusted significantly; If the evaluation score of the monocular camera is greater than the second preset score threshold and less than the first preset score threshold, output a third preset prompt, where the third preset prompt is used to indicate that the monocular camera needs to be fine-tuned.

5. The method according to claim 1, wherein The method includes: Before adjusting the monocular camera, record the historical evaluation score of the monocular camera; After adjusting the monocular camera, obtain the current evaluation score of the monocular camera; According to the current evaluation score and the historical evaluation score, determine the predicted adjustment angle of the monocular camera, and the predicted adjustment angle is used to provide a reference adjustment angle for the monocular camera.

6. The method according to claim 1, wherein The vehicle reference position includes at least one of the vehicle front end, the vehicle rear end, the vehicle side front, and the lane line position.

7. An evaluation device for the placement position of a monocular camera, characterized in that, Including: A captured image reading unit for obtaining a captured image of a single monocular camera, where the captured image includes a reference object set at a vehicle reference position; A processing unit for obtaining the image data of the captured image of the single monocular camera and the size data of the reference object, and based on the image data of the captured image and the size data of the reference object, determining the pixel accuracy value of the captured image; It includes: obtaining the characteristic edges of the reference object in the captured image according to the image data, and determining the pixel length of the characteristic edges; determining the pixel accuracy value of the reference object according to the pixel length of the characteristic edges and the dimension data of the reference object; An evaluation unit for determining the evaluation score of the captured image according to a preset evaluation quantization formula and the pixel accuracy value; including, obtaining the pixel accuracy value of the characteristic edges, and obtaining the evaluation score of the reference object according to the characteristic edges and the preset evaluation quantization formula; obtaining the average value of the evaluation scores between the reference objects according to the evaluation score of the reference object and the number of reference objects, and the average value of the evaluation scores is the evaluation score of the monocular camera; Outputting a corresponding adjustment amplitude prompt according to the evaluation score of the image captured by a single monocular camera.

8. An electronic device, characterized in that, It includes: A processor; And A memory, on which executable code is stored, and when the executable code is executed by the processor, the processor executes the method according to any one of claims 1-6.

9. A computer-readable storage medium, on which executable code is stored, and when the executable code is executed by the processor of an electronic device, the processor executes the method according to any one of claims 1-6.

Images