Vehicle paint film detection method and device, electronic equipment, storage medium and program product

Abstract

Embodiments of the present disclosure provide a paint film detection method and device for a vehicle, an electronic device, a storage medium and a program product. The method comprises: obtaining positioning point cloud data of a vehicle to be detected in a paint film detection cabin, the positioning point cloud data being three-dimensional radar point cloud data; positioning the vehicle to be detected based on the positioning point cloud data to obtain an original pose of the vehicle to be detected; performing pose offset correction on the vehicle to be detected according to the original pose and a target pose; and performing paint film detection on the vehicle to be detected after the pose offset correction is completed to obtain a paint film detection result. The above technical solution can improve the accuracy of the parking position of the vehicle to be detected, and further improve the accuracy of the paint film detection result of the vehicle.

Description

Technical Field

[0001] This disclosure relates to the field of testing technology, and more particularly to a method, apparatus, electronic device, storage medium, and program product for testing the paint film of a vehicle. Background Technology

[0002] Currently, vehicles are typically driven into a paint film inspection chamber for paint film testing. However, the accuracy of existing paint film inspection results is relatively low. Summary of the Invention

[0003] This disclosure provides a method, apparatus, electronic device, storage medium, and program product for inspecting vehicle paint film, thereby improving the accuracy of vehicle paint film inspection results.

[0004] In a first aspect, embodiments of this disclosure provide a method for detecting the paint film of a vehicle, comprising:

[0005] Acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data;

[0006] The vehicle to be detected is located based on the location point cloud data to obtain the original pose of the vehicle to be detected.

[0007] The vehicle to be detected is subjected to pose offset correction based on the original pose and the target pose.

[0008] After the pose offset correction is completed, the paint film of the vehicle to be tested is inspected to obtain the paint film inspection results.

[0009] Secondly, embodiments of this disclosure also provide a vehicle paint film inspection device, comprising:

[0010] The data acquisition module is used to acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data;

[0011] The positioning module is used to locate the vehicle to be detected based on the positioning point cloud data, so as to obtain the original pose of the vehicle to be detected.

[0012] The correction module is used to correct the pose offset of the vehicle to be detected based on the original pose and the target pose.

[0013] The paint film detection module is used to perform paint film detection on the vehicle under test after the pose offset correction is completed, and obtain the paint film detection result.

[0014] Thirdly, embodiments of this disclosure also provide an electronic device, including:

[0015] One or more processors;

[0016] Memory, used to store one or more programs.

[0017] When the one or more programs are executed by the one or more processors, the one or more processors implement the vehicle paint film detection method as described in the embodiments of this disclosure.

[0018] Fourthly, embodiments of this disclosure also provide a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements the vehicle paint film detection method as described in embodiments of this disclosure.

[0019] Fifthly, embodiments of this disclosure also provide a computer program product that, when executed by a computer, enables the computer to implement the vehicle paint film detection method as described in embodiments of this disclosure.

[0020] The vehicle paint film detection method, apparatus, electronic device, storage medium, and program product provided in this disclosure acquire positioning point cloud data of the vehicle to be detected in a paint film detection chamber. This positioning point cloud data is 3D radar data. Based on this positioning point cloud data, the vehicle to be detected is located to obtain its original pose. Based on the original pose and the target pose, pose offset correction is performed on the vehicle to be detected. After the pose offset correction is completed, paint film detection is performed on the vehicle to be detected to obtain the paint film detection result. This disclosure utilizes the above technical solution to locate the vehicle to be detected based on 3D radar point cloud data and to perform pose offset correction based on the positioning result, which can improve the accuracy of the vehicle's parking position and thus improve the accuracy of the vehicle paint film detection result. Attached Figure Description

[0021] The above and other features, advantages, and aspects of the embodiments of this disclosure will become more apparent from the accompanying drawings and the following detailed description. Throughout the drawings, the same or similar reference numerals denote the same or similar elements. It should be understood that the drawings are schematic, and the originals and elements are not necessarily drawn to scale.

[0022] Figure 1 This is a schematic flowchart of a vehicle paint film inspection method provided in an embodiment of the present disclosure;

[0023] Figure 2 A schematic diagram illustrating an optional paint film inspection process provided in an embodiment of this disclosure;

[0024] Figure 3 This is a schematic diagram illustrating another optional paint film inspection process provided in an embodiment of this disclosure;

[0025] Figure 4This is a schematic diagram of the control process of a robotic arm provided in an embodiment of the present disclosure;

[0026] Figure 5 A schematic flowchart illustrating another method for inspecting the paint film of a vehicle provided in this embodiment of the present disclosure;

[0027] Figure 6 A top view of pre-filtered localization point cloud data provided in an embodiment of this disclosure;

[0028] Figure 7 A top view of height-filtered localization point cloud data provided in an embodiment of this disclosure;

[0029] Figure 8 A top view of distance-filtered location point cloud data provided in an embodiment of this disclosure;

[0030] Figure 9 A top view of clustered and filtered localized point cloud data provided in an embodiment of this disclosure;

[0031] Figure 10 A schematic diagram showing the positions of a first dividing line and a second dividing line provided in an embodiment of this disclosure;

[0032] Figure 11 A rectangular frame diagram of a vehicle to be inspected provided in an embodiment of this disclosure;

[0033] Figure 12 A structural block diagram of a vehicle paint film inspection device provided in an embodiment of this disclosure;

[0034] Figure 13 This is a schematic diagram of the structure of an electronic device provided in an embodiment of this disclosure. Detailed Implementation

[0035] Embodiments of this disclosure will now be described in more detail with reference to the accompanying drawings. While some embodiments of this disclosure are shown in the drawings, it should be understood that this disclosure can be implemented in various forms and should not be construed as limited to the embodiments set forth herein. Rather, these embodiments are provided to provide a more thorough and complete understanding of this disclosure. It should be understood that the accompanying drawings and embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of protection of this disclosure.

[0036] It should be understood that the steps described in the method embodiments of this disclosure may be performed in different orders and / or in parallel. Furthermore, the method embodiments may include additional steps and / or omit the steps shown. The scope of this disclosure is not limited in this respect.

[0037] The term "comprising" and its variations as used herein are open-ended inclusions, meaning "including but not limited to". The term "based on" means "at least partially based on". The term "one embodiment" means "at least one embodiment"; the term "another embodiment" means "at least one additional embodiment"; the term "some embodiments" means "at least some embodiments". Definitions of other terms will be given in the description below.

[0038] It should be noted that the concepts of "first" and "second" mentioned in this disclosure are used only to distinguish different devices, modules or units, and are not used to limit the order of functions performed by these devices, modules or units or their interdependencies.

[0039] It should be noted that the terms "a" and "a plurality of" used in this disclosure are illustrative rather than restrictive, and those skilled in the art should understand that, unless otherwise expressly indicated in the context, they should be understood as "one or more".

[0040] The names of messages or information exchanged between multiple devices in the embodiments of this disclosure are for illustrative purposes only and are not intended to limit the scope of such messages or information.

[0041] It is understood that the data involved in this technical solution (including but not limited to the data itself, the acquisition or use of the data) shall comply with the requirements of relevant laws, regulations and related provisions.

[0042] Figure 1 This is a schematic flowchart illustrating a method for inspecting the paint film of a vehicle according to an embodiment of this disclosure. The method can be executed by a vehicle paint film inspection device, which can be implemented in software and / or hardware and can be configured in an electronic device, typically a paint film inspection control device, such as a computer, tablet computer, or mobile phone. The vehicle paint film inspection method provided in this disclosure is applicable to scenarios involving paint film inspection of vehicles, such as the inspection of paint film on used cars.

[0043] In related technologies, vehicles can be driven into a paint film testing chamber for paint film testing. Since the vehicle is driven into the testing chamber by the test personnel, if there are no requirements for the vehicle's parking position, it will lead to inconsistent parking positions, resulting in situations where the paint film thickness cannot be measured in some areas; if the requirements for the vehicle's parking position are too strict, it will be difficult for the test personnel to meet these requirements.

[0044] Therefore, this embodiment provides a method for inspecting the paint film of a vehicle. A three-dimensional radar is configured in the paint film inspection chamber to assist in the positioning of the vehicle entering the paint film inspection chamber, helping the tester to park the vehicle at a preset position on the turntable, thereby reducing the difficulty of parking the vehicle to be inspected and improving the accuracy of the paint film inspection results of the vehicle to be inspected.

[0045] The following describes the relevant equipment in the paint film inspection chamber of this embodiment. Multiple 3D radars and multiple cameras can be configured around the paint film inspection chamber. For example, four 3D radars and four cameras can be configured around the paint film inspection chamber. The 3D radars assist in the positioning of the vehicle to be inspected within the chamber, ensuring that the vehicle can be parked at the center point of the paint film inspection turntable, thus guaranteeing the feasibility of paint film inspection. The 3D radars and cameras can also assist in the 3D modeling of the vehicle to be inspected, enabling the output of inspection points. The paint film inspection turntable can rotate in the horizontal plane to adjust the angle of the vehicle to be inspected.

[0046] In addition, the paint film inspection chamber can also be equipped with an Automated Guided Vehicle (AGV), a robotic arm, and a paint film analyzer. The AGV is used to lift the robotic arm so that it can reach the measurable points; the robotic arm is used to grasp the paint film analyzer and perform automated paint film inspection according to the paint film inspection points; the paint film analyzer can be used to detect the paint film thickness.

[0047] Paint film inspection equipment, aided by a paint film inspection turntable, 3D radar, camera, automated guided vehicle (AGV), robotic arm, and paint film analyzer, can perform paint film inspection on vehicles. For example, the inspection turntable, 3D radar, camera, AGV, robotic arm, and paint film analyzer can all communicate with the paint film inspection equipment. Thus, when needed, the equipment can control the rotation of the inspection turntable to adjust the angle of the vehicle, control the 3D radar and camera for data acquisition, control the movement of the AGV, control the robotic arm to adjust the position of the paint film analyzer, and control the paint film analyzer to perform paint film inspection, etc.

[0048] like Figure 1 As shown, the vehicle paint film detection method provided in this embodiment may include:

[0049] S101. Obtain the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data.

[0050] The paint film inspection chamber can be understood as an inspection chamber used for paint film inspection. The vehicle to be inspected can be the vehicle currently undergoing paint film inspection, which can be a vehicle within the paint film inspection chamber, such as a vehicle currently entering the chamber or a vehicle currently positioned on the paint film inspection turntable within the chamber. For example, the vehicle to be inspected can be a vehicle awaiting shipment or a used car; the following explanation uses a used car as an example. The positioning point cloud data can be understood as point cloud data used to assist in the positioning of the vehicle to be inspected. The positioning point cloud data is 3D radar data, which can be acquired by the 3D radar within the paint film inspection chamber.

[0051] In this embodiment, the positioning point cloud data of the vehicle to be tested can be obtained to correct the pose offset of the vehicle to be tested.

[0052] Specifically, the positioning point cloud data of the vehicle to be inspected can be acquired in real time during the process of the vehicle moving towards the paint film inspection turntable or after the vehicle has been moved into the paint film inspection turntable. Based on the positioning point cloud data, the vehicle to be inspected can be corrected for positional offset in real time, so that the vehicle to be inspected can be parked on the paint film inspection turntable in a relatively ideal position (such as the offset relative to the target position is within a preset range) for paint film inspection.

[0053] The location point cloud data of the vehicle to be inspected can be collected by the 3D radar in the paint film inspection chamber. The number of 3D radars can preferably be multiple. These multiple 3D radars can be deployed around the paint film inspection chamber, such as around the center of the paint film inspection turntable, to further improve the positioning accuracy of the vehicle to be inspected.

[0054] Compared to the method of using laser ranging to locate the vehicle to be detected, laser ranging is prone to misalignment because it only has one point and has a large error for black cars and shiny cars. In contrast, 3D radar has more point cloud data, which can also provide better positioning for black cars and shiny cars. Moreover, 3D radar has better compatibility and can detect tall cars, low cars, and cars with spare tires on their backs, thus providing better positioning.

[0055] Compared to the solution of using a single radar to collect the positioning point cloud data of the vehicle under test, since a single radar can only detect point cloud data of two sides of the vehicle at most, and may only be able to detect point cloud data of one side of the vehicle when the vehicle is parked at an extreme angle, using multiple 3D radars to collect point cloud data can realize the collection of point cloud data of different sides of the vehicle under test, improve the comprehensiveness of the collected point cloud data, and thus improve the accuracy of the positioning of the vehicle under test.

[0056] Optionally, multiple three-dimensional radars are deployed in the paint film inspection chamber. The step of acquiring the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber includes: controlling the multiple three-dimensional radars to collect the original point cloud data of the vehicle to be inspected, and performing coordinate transformation on the original point cloud data to obtain the positioning point cloud data of the vehicle to be inspected in the target coordinate system.

[0057] Specifically, multiple 3D radars deployed in the paint film inspection chamber can collect 3D point cloud data of the vehicle to be inspected. For example, a data acquisition command can be sent to each of the 3D radars deployed in the paint film inspection chamber, instructing each 3D radar to collect radar point cloud data, and receiving the original point cloud data of the vehicle to be inspected returned by each 3D radar in response to the data acquisition command.

[0058] Because the 3D radars are installed in different locations and their coordinates are independent (each radar has its own radar coordinate system), the data returned is the raw point cloud data of the vehicle under test in its own radar coordinate system. Therefore, after receiving the 3D radar data returned by each radar, coordinate transformation can be performed on the received raw point cloud data to convert all the raw point cloud data returned by different 3D radars into the same coordinate system (i.e., the target coordinate system), thus obtaining the positioning point cloud data of the vehicle under test.

[0059] The target coordinate system can be set as needed; for example, it can be the world coordinate system. The raw point cloud data can be transformed into the target coordinate system using a transformation matrix, which can be obtained in advance through calibration. The transformation matrices between the radar coordinate systems and the target coordinate system of different 3D radars can be different. Therefore, when performing coordinate transformation on the raw point cloud data acquired by a certain 3D radar, for each point in the raw point cloud data acquired by that 3D radar, multiplying the coordinates of that point by the transformation matrix corresponding to that 3D radar yields the coordinates of that point in the target coordinate system.

[0060] S102. Based on the positioning point cloud data, the vehicle to be detected is located to obtain the original pose of the vehicle to be detected.

[0061] In this embodiment, after obtaining the location point cloud data of the vehicle to be tested, the vehicle to be tested can be located based on the obtained location point cloud data, such as performing pose detection on the vehicle to be tested to obtain the original pose of the vehicle to be tested.

[0062] Here, the original pose can be understood as the pose of the vehicle to be detected before pose offset correction, which may include the position and angle of the vehicle to be detected. The pose detection method of the vehicle to be detected can be flexibly selected, and this embodiment does not limit it. For example, a three-dimensional model of the vehicle to be detected can be constructed based on the positioning point cloud data of the vehicle to be detected, and the original pose of the vehicle to be detected can be determined by the position and angle of the three-dimensional model in the target coordinate system, and so on.

[0063] S103. Correct the pose offset of the vehicle to be detected based on the original pose and the target pose.

[0064] In this embodiment, after obtaining the original pose of the vehicle to be inspected, pose offset correction can be performed on the vehicle to be inspected based on the original pose and the target pose. For example, the offset amount of the original pose relative to the target pose can be determined, and the vehicle to be inspected can be moved and / or the paint film inspection turntable can be rotated according to the offset amount to correct the offset amount and adjust the vehicle to be inspected from the original pose to the target pose. Here, the target pose can be understood as the desired pose of the vehicle to be inspected, which can be set in advance as needed.

[0065] In some implementations, the offset can be corrected only when the offset between the original pose and the target pose of the vehicle under test is greater than a preset threshold, thereby improving the overall paint film detection speed of the vehicle under test. Optionally, the vehicle under test is located on a paint film detection turntable. The pose offset correction of the vehicle under test based on the original pose and the target pose includes: determining the distance offset and angle offset of the vehicle under test based on the original pose and the target pose; if the distance offset is less than or equal to a preset distance threshold and the angle offset is less than or equal to a preset angle threshold, then the pose offset correction of the vehicle under test is determined to be complete; if the distance offset is greater than the preset distance threshold, a prompt message is displayed, which prompts the control of the vehicle under test to move to a target distance corresponding to the distance offset; if the angle offset is greater than a preset angle threshold, then the control of the paint film detection turntable to rotate to a target angle corresponding to the angle offset is established.

[0066] For example, the distance offset and angle offset of the original pose relative to the target pose can be determined based on the original pose and the target pose.

[0067] If the distance offset is less than or equal to the preset distance threshold and the angle offset is less than or equal to the preset angle threshold, then there is no need to correct the pose offset of the vehicle to be detected, and the pose offset correction of the vehicle to be detected is completed.

[0068] If the distance offset exceeds a preset distance threshold, a prompt message will be displayed regardless of whether the angle offset exceeds a preset angle threshold. This prompt message will instruct the tester to move the vehicle under test a target distance corresponding to the distance offset, thereby correcting the offset. The tester can then use this prompt message to drive the vehicle under test a target distance.

[0069] If the angular offset is greater than a preset angle threshold, the paint film detection turntable can be rotated to the target angle corresponding to this angular offset, regardless of whether the distance offset is greater than the preset distance threshold, in order to correct the angular offset. There are no restrictions on when to correct the angular offset; for example, it can be corrected before or after correcting the distance offset, or when a correction instruction is received from the tester. The specific timing can be flexibly set according to needs.

[0070] The target distance can be determined based on the distance offset of the vehicle to be detected, and is used to correct for the distance offset of the vehicle to be detected. The target angle can be determined based on the angle offset of the vehicle to be detected, and is used to correct for the angle offset of the vehicle to be detected.

[0071] S104. After the pose offset correction is completed, the paint film of the vehicle to be tested is inspected to obtain the paint film inspection result.

[0072] In this embodiment, after the vehicle's pose offset is corrected, the paint film can be inspected, for example, using a paint film analyzer. The type of paint film analyzer is not limited; for example, it can be a non-contact model. Exemplarily, the paint film analyzer can use a photothermal method to detect the paint film thickness of the vehicle.

[0073] In some implementations, considering that different vehicles to be inspected may have different models, especially in the scenario of paint film inspection of used cars, the different models of the vehicles to be inspected are usually different. Therefore, a 3D model of the vehicle to be inspected can be created, and the inspection points of the vehicle to be inspected can be determined based on the 3D model of the vehicle to be inspected, so as to improve the accuracy of the determined inspection points and thus improve the accuracy of the paint film inspection results.

[0074] Optionally, the step of performing paint film detection on the vehicle to be inspected includes: acquiring detection point cloud data and image data of the vehicle to be inspected, wherein the detection point cloud data is three-dimensional radar point cloud data; performing three-dimensional modeling of the vehicle to be inspected based on the detection point cloud data and the image data to obtain a three-dimensional model of the vehicle to be inspected; determining the detection point information of the vehicle to be inspected based on the three-dimensional model; and controlling the paint film analyzer to perform paint film detection on the vehicle to be inspected based on the detection point information.

[0075] The detection point cloud data can be used to construct a 3D model of the vehicle under inspection for paint film inspection. This point cloud data can be acquired by a 3D radar within the inspection chamber. Image data of the vehicle under inspection can be obtained from cameras within the paint film inspection chamber.

[0076] For example, such as Figure 2 As shown, inspectors can send paint film inspection instructions (i.e., issue paint film inspection tasks) to the paint film inspection control equipment (such as a central control unit, which can be an edge gateway) for the vehicle to be inspected via a mobile terminal and / or a server.

[0077] Therefore, after receiving the paint film inspection command, the paint film inspection control equipment can control the various 3D radars and cameras in the paint film inspection chamber to collect data on the vehicle under inspection. Based on the radar point cloud data collected by each 3D radar, it obtains the inspection point cloud data of the vehicle under inspection in the target coordinate system. Based on the image data collected by each camera, it performs component segmentation on the vehicle under inspection, generating component segmentation results. Based on the inspection point cloud data and the component segmentation results, it performs 3D modeling of the vehicle under inspection, obtaining a 3D model of the vehicle under inspection. Based on the 3D model of the vehicle under inspection, it determines coordinate data, such as the inspection point information. Based on this coordinate data, it controls the paint film analyzer to perform paint film inspection on the vehicle under inspection, obtaining the paint film thickness of the vehicle under inspection at each inspection point, and generating the paint film inspection results of the vehicle under inspection based on the paint film thickness at each inspection point.

[0078] The detection point information can be the position coordinates of the detection point within the target coordinate system. These detection points can be understood as the locations on the vehicle to be inspected where paint film inspection is required. The 3D modeling method used for the vehicle to be inspected is not limited and can be flexibly set according to needs.

[0079] Optionally, the paint film analyzer is mounted on a robotic arm, which is mounted on an automated guided vehicle (AGV). Determining the detection point information of the vehicle to be inspected based on the 3D model includes: determining the target point information of the AGV and at least one detection point information corresponding to the target point information based on the 3D model; controlling the paint film analyzer to perform paint film inspection on the vehicle to be inspected according to the detection point information includes: controlling the AGV and the robotic arm to move the paint film analyzer to the detection point corresponding to the detection point information according to the target point information and the detection point information; and controlling the paint film analyzer to perform paint film inspection on the detection point.

[0080] The target point information can be the position coordinates of the target point to which the automated guided vehicle (AGV) needs to move within the target coordinate system. One target point information can correspond to one or more detection point information. That is, after the AGV moves to a certain target point, it can control the paint thickness gauge to measure the paint thickness of the vehicle under test at one or more detection points.

[0081] Specifically, the target point information of the automated guided vehicle (AGV) and at least one detection point information corresponding to each target point information can be determined based on a 3D model. According to the target point information and the detection point information, the AGV and the robotic arm are controlled to move the paint film analyzer to the detection point corresponding to the detection point information, and the paint film analyzer is controlled to perform paint film detection at the detection point.

[0082] For example, based on the detection order of each target point information, the target point corresponding to each target point information can be sequentially designated as the current target point, instructing the automated guided vehicle (AGV) to move to the current target point. After the AGV moves to the current target point, based on the detection order of each detection point information corresponding to the current target point, the detection point corresponding to each detection point information can be sequentially designated as the current detection point, instructing the robotic arm to move the paint film analyzer to the current detection point, instructing the paint film analyzer to perform paint film detection at the current detection point, and, after receiving the paint film thickness returned by the paint film analyzer at the current detection point, designating the next detection point as the current detection point, and returning to instruct the robotic arm to move the paint film analyzer to the current detection point, until all detection points corresponding to the current target point have been detected. When all detection points corresponding to the current target point have been detected, the next target point is designated as the current target point, and returning to instruct the AGV to move to the current target point, until there are no more next target points.

[0083] Optionally, based on the target location information and the detection location information, the automated guided vehicle and the robotic arm are controlled to move the paint film analyzer to the detection location corresponding to the detection location information, including: controlling the automated guided vehicle to move to the target location corresponding to the target location information; and after the automated guided vehicle moves to the target location, controlling the robotic arm to move the paint film analyzer to the detection location corresponding to the detection location information.

[0084] As an optional implementation, the target point information and detection point information determined based on the three-dimensional model can be all target point data and all detection point data of the vehicle to be inspected. In this case, the paint film detection turntable does not need to be controlled to rotate during the inspection process. For example, the automatic guide vehicle can be controlled to move around the vehicle to be inspected to perform paint film inspection.

[0085] As an alternative implementation, the target point data and detection point data determined based on the 3D model can be the target point data and detection point data of the target side of the vehicle to be inspected facing the target direction. In this case, the paint film detection turntable can be controlled to rotate during the inspection process. For example, after the inspection of one side is completed, the paint film detection turntable can be controlled to rotate a certain angle (such as 90 degrees), and the 3D model of the vehicle to be inspected can be corrected based on the rotated detection point cloud data and image data. Based on the corrected 3D model, the target point data and detection point data of the target side of the vehicle to be inspected facing the target direction after rotation are output. This process is repeated to achieve paint film inspection of the vehicle to be inspected. At this time, the automated guided vehicle can be controlled to move only on the side corresponding to the target direction.

[0086] For example, such as Figure 3 As shown, the paint film inspection process can be described as follows:

[0087] The server sends paint film detection instructions to the central control unit (i.e., the paint film detection control equipment) for the vehicle to be inspected.

[0088] After receiving the paint film detection command, the central control unit controls the 3D radar in the paint film detection chamber to collect radar data (i.e., detection point cloud data) of the vehicle to be detected, and controls the camera in the paint film detection chamber to collect image data of the vehicle to be detected by taking pictures.

[0089] The central control unit performs component segmentation on the vehicle under test based on the image data of the vehicle under test. Based on the component segmentation results and the radar data of the vehicle under test, a 3D model of the vehicle under test is created, resulting in a 3D model of the vehicle under test. Based on the 3D model of the vehicle under test, the coordinate points (i.e., target point information) and measurement points (i.e., detection point information) of the AGV when the first side of the vehicle under test is detected are obtained.

[0090] The central control unit controls the robotic arm to switch to a first posture, and controls the AGV to drive the robotic arm into the measurement area. The coordinate transformation of the robotic arm is calculated, and measurement begins. After measurement in the first posture is completed, the AGV drives the robotic arm back to the safe zone. After exiting the safe zone, the robotic arm switches to a second posture, and the AGV drives the robotic arm into the measurement area again. The coordinate transformation of the robotic arm is calculated, and measurement begins. After measurement in the first posture is completed, the AGV drives the robotic arm back to the safe zone. At this point, the paint film inspection of the first side of the vehicle under inspection is complete. The first and second postures can be different postures used for paint film inspection of different parts of the vehicle. For example, when the robotic arm is in the first posture, paint film inspection can be performed on the top of the vehicle; when it is in the second posture, paint film inspection can be performed on the side. Or, when the robotic arm is in the first posture, paint film inspection can be performed on the side; when it is in the second posture, paint film inspection can be performed on the top, and so on. Controlling the robotic arm to retreat to the safe zone before switching postures avoids collisions with the vehicle under inspection during the posture transition process.

[0091] After the paint film on the first side of the vehicle to be inspected is inspected, the central control unit controls the paint film inspection turntable to rotate 90 degrees in a preset direction (such as clockwise or counterclockwise).

[0092] After the paint film inspection turntable has finished rotating, the central control unit restarts control of the 3D radar in the paint film inspection chamber to collect radar data of the vehicle to be inspected, and controls the camera in the paint film inspection chamber to collect image data of the vehicle to be inspected by taking pictures.

[0093] The central control unit performs component segmentation on the vehicle under test based on the re-acquired image data. Based on the component segmentation results and the re-acquired radar data, it verifies the 3D model of the vehicle under test. If the verification passes, it obtains the AGV coordinates and measurement points for inspecting the second side of the vehicle based on the 3D model and re-executes the aforementioned robotic arm control process to complete the paint film inspection on the second side of the vehicle. Alternatively, if the verification fails, it re-models the vehicle under test based on the component segmentation results and the re-acquired radar data, obtains the AGV coordinates and measurement points for inspecting the second side of the vehicle based on the re-modeled 3D model, and re-executes the aforementioned robotic arm control process to complete the paint film inspection on the second side of the vehicle.

[0094] This process is repeated four times to complete the paint film inspection of the vehicle under inspection. After the measurement is completed, the AGV can be controlled to move the robotic arm back to the fully equipped area and return the measurement results (i.e., the paint film inspection results) of the vehicle under inspection to the server.

[0095] In some examples, such as Figure 4As shown, the control process of the robotic arm can be described as follows:

[0096] The central control unit acquires the current measurement data (i.e., detection point information) of the vehicle to be tested, determines the current measurement data according to the detection order of each measurement data, and sends it to the robotic arm.

[0097] The robotic arm receives the current measurement data, moves accordingly (e.g., performs pose transformation), and determines whether to trigger protection based on detection data or status from torque sensors, distance sensors, and / or microswitches during movement. If protection is triggered, it stays at the current position and returns a detection failure message to the central control unit. If protection is not triggered, it moves to the measurement position corresponding to the current measurement data.

[0098] After the robotic arm moves to the measurement position corresponding to the current measurement data, the central control unit triggers the paint thickness gauge to measure, such as sending a measurement command to the paint thickness gauge to instruct it to measure the paint thickness of the vehicle under test and return the measurement data at this measurement position.

[0099] The central control unit determines whether the paint thickness gauge has successfully returned measurement data. If the paint thickness gauge successfully returns measurement data, it checks whether there is a next measurement data to be taken on the current side. If there is a next measurement data to be taken, it takes the next measurement data as the current measurement data and returns to execute the operation of sending it to the robotic arm, until there is no next measurement data to be taken. If there is no next measurement data to be taken, it further checks whether the entire vehicle under test has been measured. If so, it uploads the measurement results of the vehicle under test to the server and can control the paint thickness detection turntable to rotate 90 degrees to restore the vehicle under test to its pre-test posture. If not, it controls the robotic arm to convert Euler angles to change the posture, controls the paint thickness detection turntable to rotate 90 degrees, takes the converted side as the current side, and returns to execute the operation of obtaining the measurement data to be taken on the current side of the vehicle under test. Furthermore, if the paint thickness gauge fails to return measurement data, it checks whether the number of retries at this measurement position is higher than a preset threshold. If the number of retries is lower than or equal to the preset threshold, it returns to re-execute the operation of triggering the paint thickness gauge measurement; if the number of retries is higher than the threshold, it executes the above operation of checking whether there is a next measurement data to be taken on the current side.

[0100] The vehicle paint film detection method provided in this embodiment acquires the positioning point cloud data of the vehicle to be detected in the paint film detection chamber. This positioning point cloud data is 3D radar data. Based on the positioning point cloud data, the vehicle to be detected is located to obtain the original pose of the vehicle. The pose of the vehicle to be detected is corrected for pose offset based on the original pose and the target pose. After the pose offset correction is completed, the paint film of the vehicle to be detected is detected to obtain the paint film detection result. This embodiment utilizes the above technical solution to locate the vehicle to be detected based on 3D radar point cloud data and corrects the pose offset of the vehicle to be detected based on the positioning result, which can improve the accuracy of the parking position of the vehicle to be detected, thereby improving the accuracy of the vehicle paint film detection result.

[0101] Figure 5 This is a schematic flowchart illustrating another method for detecting vehicle paint film provided in this embodiment. The solution in this embodiment can be combined with one or more optional solutions in the above embodiments. Optionally, the step of locating the vehicle to be detected based on the positioning point cloud data to obtain the original pose of the vehicle to be detected includes: performing two-dimensional mapping on the positioning point cloud data to obtain target point cloud data of the vehicle to be detected; determining the baseline of the vehicle to be detected based on the target point cloud data; constructing a rectangular frame of the vehicle to be detected based on the baseline; and determining the original pose of the vehicle to be detected based on the rectangular frame.

[0102] Correspondingly, such as Figure 5 As shown, the vehicle paint film detection method provided in this embodiment may include:

[0103] S201. Obtain the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data.

[0104] S202. Perform two-dimensional mapping on the positioning point cloud data to obtain the target point cloud data of the vehicle to be detected.

[0105] In this implementation, the location point cloud data can be mapped in two dimensions, and the vehicle to be detected can be located based on the mapped two-dimensional point cloud data, so as to simplify the amount of calculation required to locate the vehicle to be detected and further improve the location speed of the vehicle to be detected.

[0106] Specifically, each point in the positioning point cloud data can be mapped onto a preset plane, such as the plane formed by the x-axis and y-axis of the target coordinate system, and the resulting point cloud data can be used as the target point cloud data for the vehicle to be detected. Here, the positioning point cloud data is three-dimensional point cloud data in the target coordinate system, and the target point cloud data is two-dimensional point cloud data obtained by two-dimensional mapping the positioning point cloud data.

[0107] In some implementations, the positioning point cloud data can be filtered to remove noise points, further improving the accuracy of the vehicle positioning result. Optionally, before performing two-dimensional mapping on the positioning point cloud data, the method further includes: performing point cloud filtering on the positioning point cloud data using a preset filtering method, wherein the preset filtering method includes at least one of height filtering, distance filtering, and clustering filtering; the two-dimensional mapping of the positioning point cloud data includes: performing two-dimensional mapping on the filtered positioning point cloud data.

[0108] The preset filtering methods can include, but are not limited to, one or more of height filtering, distance filtering, and clustering filtering, which can be set as needed. Height filtering can be understood as filtering based on the height of each point in the positioning point cloud data, filtering out points in the positioning point cloud data whose height is greater than a set height threshold; distance filtering can be understood as filtering based on the distance between each point in the positioning point cloud and a preset position (such as the center of the paint film detection turntable), filtering out points in the positioning point cloud data whose distance from the preset position is greater than a set distance; clustering filtering can be understood as dividing the points in the positioning point cloud into clusters and filtering out points in smaller clusters, such as keeping only the cluster containing the most points (i.e., the largest cluster) and deleting other clusters.

[0109] Specifically, after obtaining the location point cloud data of the vehicle to be detected, point cloud filtering can be performed using methods such as height filtering, distance filtering, and / or cluster filtering. After filtering, the filtered location point cloud data can be mapped in two dimensions to obtain the target point cloud data of the vehicle to be detected.

[0110] Figure 6 This is a top view of pre-filtered positioning point cloud data provided in an embodiment of this disclosure (points other than black dots are auxiliary points, and lines other than black lines are auxiliary lines; auxiliary points and lines may not be included in the positioning point cloud data; auxiliary points and / or lines not mentioned in this embodiment can be considered as auxiliary points and / or lines not involved in the vehicle positioning process), such as Figure 6 As shown, the unfiltered point cloud data contains many noisy points. Taking preset filtering methods including height filtering, distance filtering, and cluster filtering as examples, the filtering process of the point cloud data can be described as follows:

[0111] A1. Perform high-level filtering on the location point cloud data.

[0112] Specifically, a set height threshold corresponding to the vehicle to be detected can be obtained, and points in the positioning point cloud whose height value is greater than the set height threshold can be deleted.

[0113] The height threshold can be set in advance or determined based on the vehicle information of the vehicle to be detected.

[0114] For example, the vehicle model of the vehicle to be detected can be determined based on the vehicle identification information; the height value of the lower boundary of the vehicle window can be determined based on the vehicle model, and this height value can be used as a set height threshold corresponding to the vehicle to be detected. Thus, points in the positioning point cloud with a height higher than the lower boundary of the window can be filtered out by the set height threshold, and only points with a height lower than the lower boundary of the window can be retained.

[0115] A2. Perform distance filtering on the height-filtered location point cloud data.

[0116] Figure 7 A top view of height-filtered localization point cloud data provided in an embodiment of this disclosure, such as... Figure 7 As shown, even after height filtering, some noise points may still exist in the positioning point cloud data, such as scanning noise points from the device (e.g., ...). Figure 7 The device shown in the lower right corner), wall noise points, and some discrete noise points, etc. Therefore, after height filtering, distance filtering can be further performed to filter out noise points that are far away from the vehicle being detected.

[0117] For example, since the vehicle to be inspected is located on the paint film inspection turntable, and the size of the paint film inspection turntable is generally larger than the length of the vehicle to be inspected, there is almost no situation where the parking error of the vehicle to be inspected is too large, causing the vehicle to be completely parked outside the turntable. Therefore, the center of the paint film inspection turntable can be used as the preset position, and a set distance greater than or equal to the distance between the boundary of the paint film inspection turntable and its center can be selected. This set distance can be preset, and its specific value is not limited. Taking the test turntable as a circle with a radius of 2.75m as an example, the set distance can be set to 3m or 3.1m, etc.

[0118] Therefore, when performing distance filtering, for each point in the height-filtered positioning point cloud data, the distance between that point and the preset position can be calculated, and it can be determined whether the distance is greater than the set distance. If so, the point is deleted from the positioning point cloud data; if not, the point is retained in the positioning point cloud data.

[0119] A3. Perform cluster filtering on the distance-filtered location point cloud data.

[0120] Figure 8 A top view of distance-filtered localization point cloud data provided in an embodiment of this disclosure, such as... Figure 8 As shown, some noisy points may still exist in the location point cloud data after distance filtering, such as people or objects placed on the paint film inspection turntable. Figure 7The items placed on the right side of the vehicle shown in the picture. Figure 6-7 (not shown in the image), etc. Therefore, after distance filtering, cluster filtering can be further performed to filter out smaller clusters in the location point cloud data.

[0121] Specifically, a preset clustering filtering algorithm can be used to cluster the distance-filtered location point cloud data, and smaller clusters can be deleted from the location point cloud data. Considering that the vehicle to be detected is generally the largest object on the paint film detection turntable, for example, only the largest cluster in the location point cloud data can be retained, that is, only the cluster with the most points can be retained, and all other clusters can be deleted. A top view of the location point cloud data after clustering and filtering is shown below. Figure 9 As shown in the figure. The preset clustering filtering algorithm can be flexibly set as needed.

[0122] S203. Determine the baseline of the vehicle to be detected based on the target point cloud data.

[0123] Considering that to obtain the original pose of the vehicle under test, i.e., to obtain the position and angle of the vehicle under test, the outer contour rectangle (i.e., the bounding box) of the vehicle under test can be obtained, and the original pose of the vehicle under test can be determined based on the bounding box. Therefore, in this embodiment, a baseline can be constructed based on the mapped target point cloud data, so that the outer contour rectangle of the vehicle under test can be obtained based on the baseline.

[0124] Specifically, a baseline for the vehicle to be detected can be drawn based on the mapped target point cloud data. The method of drawing the baseline can be set as needed. For example, a target straight line can be drawn based on the target point cloud data, such that the target straight line passes through at least two points in the target point cloud data, and the target straight line is the line with the most associated points in the target point cloud. After obtaining a target straight line that meets the above conditions, this target straight line can be used as the baseline for the vehicle to be detected.

[0125] In some implementations, the baseline of the vehicle to be detected can be determined according to the point cloud region to further reduce the computational load required for the baseline determination process. Optionally, determining the baseline of the vehicle to be detected based on the target point cloud data includes: drawing a first dividing line based on the target pose, dividing the target point cloud data into a first point cloud region and a second point cloud region through the first dividing line, wherein the first dividing line is parallel to the vehicle body direction corresponding to the target pose and passes through the target feature points of the target point cloud data; drawing the baseline of the vehicle to be detected based on the first point cloud region or the second point cloud region, wherein the associated points of the baseline in the corresponding point cloud region satisfy preset conditions, and the distance between the associated points of the baseline and the baseline is within a preset distance range.

[0126] The first dividing line can be understood as a dividing line parallel to the vehicle body direction corresponding to the target pose and passing through the target feature points. The target feature points can be preset feature points of the target point cloud data, such as the center point or centroid of the target point cloud data. The following explanation uses the centroid of the target point cloud data as an example. The first point cloud region and the second point cloud region can be two point cloud regions of the target point cloud data obtained by dividing them through the first dividing line. The baseline can also be understood as a standard line. The baseline can be used as a reference for constructing the rectangular frame of the vehicle to be detected; that is, the rectangular frame of the vehicle to be detected can be constructed using the baseline as a reference.

[0127] The associated points of a certain straight line (such as a baseline, first straight line, second straight line, third straight line, fourth straight line, or fifth straight line, etc.) can be understood as points in the target point cloud data or the corresponding point cloud region that are within a preset distance range from that straight line. Considering that the coordinates of points in the point cloud data have a certain degree of error, a preset distance range can be set as an acceptable error. When the point is within the preset distance range, it can be considered that the point is located on the straight line. In other words, at this time, the associated points of the straight line can be regarded as points in the target point cloud data that are located on the straight line. The preset distance range can be set as needed, such as 2mm, 5mm, or 10mm, etc.

[0128] The preset conditions can be conditions associated with the associated points, such as conditions for limiting the proportion of associated points in the corresponding point cloud region and / or conditions for limiting the number of associated points. For example, the preset conditions may include a first preset condition and / or a second preset condition, wherein the first preset condition is associated with the proportion of associated points in the corresponding point cloud region and the second preset condition is associated with the number of associated points.

[0129] For example, target feature points of target point cloud data can be determined, such as by calculating the centroid of the target point cloud data based on the coordinates of each point in the target point cloud data, and using it as the target feature point of the target point cloud data.

[0130] After obtaining the target feature points 100 of the target point cloud data, a straight line parallel to the vehicle body direction corresponding to the target pose can be drawn through the target feature points 100 as the first dividing line 101. The target point cloud data is then divided into a first point cloud region and a second point cloud region using the first dividing line 101. For example, the target point cloud data can be divided into a left region located to the left of the first dividing line 101 and a right region located to the right of the first dividing line 101. Figure 10 As stated above.

[0131] After obtaining the first and second point cloud regions, a straight line that satisfies preset conditions based on the associated points of either region can be used as the baseline for the vehicle to be inspected. For example, a straight line can be found in the first point cloud region such that it has the most associated points compared to other straight lines in the first point cloud region, and this line can be used as the baseline; or, a straight line can be found in the second point cloud region such that it has the most associated points compared to other straight lines in the second point cloud region, and this line can be used as the baseline, and so on.

[0132] Optionally, the preset conditions include a first preset condition and / or a second preset condition. The step of drawing the baseline of the vehicle to be tested based on the first point cloud region or the second point cloud region includes: obtaining a first straight line in the first point cloud region whose proportion of associated points meets the first preset condition; if the first straight line is obtained, then the first straight line is used as the baseline of the vehicle to be tested; if the first straight line is not obtained, then obtaining second straight lines in the first point cloud region and the second point cloud region respectively whose number of associated points meets the second preset condition, and selecting a second straight line as the baseline of the vehicle to be tested based on the number of associated points of each second straight line.

[0133] For example, a first straight line whose proportion of associated points meets a first preset condition can be obtained first in the first point cloud region.

[0134] For example, a straight line can be formed by any two points in the first point cloud region. The algorithm then determines whether the proportion of the associated points of this straight line within the total number of points in the first point cloud region meets a first preset condition, such as a preset proportion threshold. If yes, this straight line is designated as the first straight line, and the process of acquiring the first straight line ends. If not, the algorithm returns to the operation of forming a straight line by any two points in the first point cloud region until the first straight line is acquired or the points in the first point cloud region have been traversed. If the first straight line is not acquired after the points in the first point cloud region have been traversed, it can be determined that the first straight line has not been acquired. It should be noted that when returning to the operation of forming a straight line by any two points in the first point cloud region, it is possible to select two points that have already been used to draw a straight line, i.e., to draw a straight line that has not yet been drawn.

[0135] If the first straight line is obtained in the first point cloud region, the first straight line is used as the baseline of the vehicle to be detected.

[0136] If the first straight line is not obtained in the first point cloud region, a second straight line that meets the second preset condition can be obtained in the first point cloud region or the second point cloud region, and used as the baseline of the vehicle to be detected.

[0137] To improve the confidence of the acquired second straight line, for example, a second straight line that meets the second preset condition can be obtained from the first point cloud region and the second point cloud region respectively. For example, a straight line with the most associated points in the corresponding point cloud region can be obtained from the first point cloud region and the second point cloud region respectively, thus obtaining two second straight lines. Then, based on the number of associated points in the point cloud regions where these two second straight lines are located, the second straight line with the most associated points is selected from these two second straight lines as the baseline of the vehicle to be detected.

[0138] S204. Construct a rectangular frame of the vehicle to be detected based on the baseline.

[0139] In this embodiment, after obtaining the baseline 102 of the vehicle to be tested (e.g., ... Figure 10 As shown in the figure (taking the baseline 102 located in the left point cloud area of ​​the target point cloud data as an example), a rectangular box of the vehicle to be detected can be constructed based on the baseline.

[0140] In some embodiments, after drawing the first segmentation line according to the target pose, the method further includes: drawing a second segmentation line perpendicular to the first segmentation line through the target feature points, and dividing the target point cloud data into a third point cloud region and a fourth point cloud region through the second segmentation line; constructing the rectangular frame of the vehicle to be detected based on the baseline includes: drawing a third straight line parallel to the baseline for the point cloud regions in the first and second point cloud regions where the baseline is not located, where the associated points meet preset conditions; drawing a fourth straight line and a fifth straight line in the third and fourth point cloud regions respectively, where the associated points meet preset conditions and are perpendicular to the baseline, and using the rectangle enclosed by the baseline, the third straight line, the fourth straight line and the fifth straight line as the rectangular frame of the vehicle to be detected.

[0141] The second dividing line can be understood as a dividing line perpendicular to the vehicle body direction corresponding to the target pose and passing through the target feature point; that is, the second dividing line passes through the target feature point and is perpendicular to the first dividing line. The third and fourth point cloud regions can be two point cloud regions of the target point cloud data obtained by dividing by the second dividing line. When the first and second point cloud regions are the left and right regions of the target point cloud data, respectively, the third and fourth point cloud regions can be the upper and lower regions of the target point cloud data, respectively.

[0142] For example, such as Figure 10As shown, a second dividing line 103 perpendicular to the first dividing line 101 can be drawn through the target feature point 100. The target point cloud data can be divided into a third point cloud region and a fourth point cloud region through the second dividing line 103. For example, the target point cloud data can be divided into an upper region above the second dividing line 103 and a lower region below the second dividing line 103 through the second dividing line 103.

[0143] Taking the baseline located in the first point cloud region as an example, the process of constructing the bounding box of the vehicle to be inspected can be described as follows: Find a third straight line in the second point cloud region such that this third straight line is parallel to the baseline, and the associated points of this third straight line in the second point cloud region satisfy preset conditions; find a fourth straight line in the third point cloud region such that this fourth straight line is perpendicular to the baseline, and the associated points of this fourth straight line in the third point cloud region satisfy preset conditions; and find a fifth straight line in the fourth point cloud region such that this fifth straight line is perpendicular to the baseline, and the associated points of this fifth straight line in the fourth point cloud region satisfy preset conditions. After obtaining the third, fourth, and fifth straight lines, the rectangle enclosed by the baseline, the third, fourth, and fifth straight lines can be used as the bounding box of the vehicle to be inspected.

[0144] As some alternative implementation methods, the third, fourth, and fifth lines can be the lines with the most associated points in the corresponding point cloud region. The search process for these lines is similar to that for the second line. For details, please refer to the search process for the second line.

[0145] Considering that some vehicles have a curved front profile, the straight lines corresponding to the front area obtained based on preset conditions may not be the exact outline of the front. Therefore, to improve the accuracy of the determined bounding boxes, for the point cloud region corresponding to the front, such as... Figure 10 The lower point cloud region shown can be used to obtain a sixth straight line that passes through at least two points in the point cloud region, is perpendicular to the baseline, and is farthest from the target feature point. The proportion of points in the point cloud region that are within a set distance from the sixth straight line is obtained out of all points included in the point cloud region. If this proportion is greater than a set threshold, the sixth straight line is used to determine the bounding box of the detected vehicle. Figure 11 As shown, the sixth straight line is used as the boundary line of the rectangular region within the point cloud region; if this ratio is less than or equal to the set ratio threshold, the above straight lines that meet the preset conditions in the point cloud region are used to determine the rectangular box of the vehicle to be detected.

[0146] The set distance range and the preset distance range can be the same or different. For example, the preset distance range can be a sub-range of the set distance range. The set ratio threshold and the preset ratio threshold can be the same or different, and can be set as needed.

[0147] S205. Determine the original pose of the vehicle to be detected based on the rectangular frame.

[0148] For example, after obtaining the bounding box of the vehicle to be detected, the original pose of the vehicle to be detected can be determined based on the bounding box. For instance, the center point of the bounding box can be used as the position coordinates of the vehicle to be detected, and the angle of the bounding box along the central axis in the vehicle body direction can be used as the angle of the vehicle to be detected. Thus, the original pose of the vehicle to be detected can be obtained.

[0149] S206. Correct the pose offset of the vehicle to be detected based on the original pose and the target pose.

[0150] S207. After the pose offset correction is completed, the paint film of the vehicle to be tested is inspected to obtain the paint film inspection result.

[0151] The vehicle paint film inspection method provided in this embodiment first determines the baseline of the vehicle to be inspected, and then constructs a rectangular bounding box of the vehicle to be inspected based on the baseline. This improves the accuracy of the determined rectangular bounding box, thereby improving the accuracy of the determined original pose of the vehicle to be inspected. In addition, by performing two-dimensional mapping on the positioning point cloud data, the computational load required for the original pose determination process can be reduced, thus increasing the speed of determining the original pose of the vehicle to be inspected.

[0152] Figure 12 This is a structural block diagram of a vehicle paint film inspection device provided in an embodiment of this disclosure. The device can be implemented by software and / or hardware and can be configured in an electronic device, typically within a paint film inspection control device, such as a computer, tablet, or mobile phone. It can perform paint film inspection on a vehicle by executing a paint film inspection method, such as for inspecting the paint film of a used car. Figure 12 As shown, the vehicle paint film detection device provided in this embodiment may include: a data acquisition module 1201, a positioning module 1202, a correction module 1203, and a paint film detection module 1204, wherein,

[0153] Data acquisition module 1201 is used to acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data;

[0154] The positioning module 1202 is used to locate the vehicle to be detected based on the positioning point cloud data to obtain the original pose of the vehicle to be detected.

[0155] Correction module 1203 is used to correct the pose offset of the vehicle to be detected based on the original pose and the target pose;

[0156] The paint film detection module 1204 is used to perform paint film detection on the vehicle to be tested after the pose offset correction is completed, and to obtain the paint film detection result.

[0157] The vehicle paint film detection device provided in this embodiment acquires the positioning point cloud data of the vehicle to be detected in the paint film detection chamber through a data acquisition module. This positioning point cloud data is three-dimensional radar data. A positioning module locates the vehicle to be detected based on this positioning point cloud data to obtain the original pose of the vehicle. A correction module corrects the pose offset of the vehicle to be detected based on the original pose and the target pose. After the pose offset correction is completed, the paint film detection module performs paint film detection on the vehicle to be detected to obtain the paint film detection result. This embodiment utilizes the above technical solution to locate the vehicle to be detected based on three-dimensional radar point cloud data and corrects the pose offset of the vehicle to be detected based on the positioning result, which can improve the accuracy of the vehicle's parking position and thus improve the accuracy of the vehicle paint film detection result.

[0158] Optionally, the positioning module 1202 includes: a mapping unit for performing two-dimensional mapping on the positioning point cloud data to obtain target point cloud data of the vehicle to be detected; a baseline determination unit for determining the baseline of the vehicle to be detected based on the target point cloud data; a rectangle construction unit for constructing a rectangle of the vehicle to be detected based on the baseline; and a pose determination unit for determining the original pose of the vehicle to be detected based on the rectangle.

[0159] Furthermore, the vehicle paint film detection device may further include: a filtering module, used to perform point cloud filtering on the positioning point cloud data using a preset filtering method before performing two-dimensional mapping on the positioning point cloud data, the preset filtering method including at least one of height filtering, distance filtering and clustering filtering; the mapping unit may be specifically used to perform two-dimensional mapping on the filtered positioning point cloud data.

[0160] Optionally, the baseline determination unit includes: a point cloud division subunit, used to draw a first dividing line according to the target pose, dividing the target point cloud data into a first point cloud region and a second point cloud region through the first dividing line, wherein the first dividing line is parallel to the vehicle body direction corresponding to the target pose and passes through the target feature points of the target point cloud data; and a baseline drawing subunit, used to draw the baseline of the vehicle to be detected based on the first point cloud region or the second point cloud region, wherein the associated points of the baseline in the corresponding point cloud region meet preset conditions, and the distance between the associated points of the baseline and the baseline is within a preset distance range.

[0161] Optionally, the preset conditions include a first preset condition and / or a second preset condition, and the baseline drawing subunit is specifically used to: obtain a first straight line in the first point cloud region whose proportion of associated points meets the first preset condition; if the first straight line is obtained, then the first straight line is used as the baseline of the vehicle to be detected; if the first straight line is not obtained, then the number of associated points in the first point cloud region and the second point cloud region respectively meets the second preset condition, and a second straight line is selected as the baseline of the vehicle to be detected based on the number of associated points of each second straight line.

[0162] Furthermore, the vehicle paint film detection device may further include: a point cloud segmentation module, used to draw a second segmentation line perpendicular to the first segmentation line through the target feature points after drawing the first segmentation line according to the target pose, and to divide the target point cloud data into a third point cloud region and a fourth point cloud region through the second segmentation line; the rectangular frame construction unit is specifically used to: draw a third straight line that satisfies the preset conditions and is parallel to the baseline for the point cloud regions in the first point cloud region and the second point cloud region where the baseline is not located; draw a straight line that satisfies the preset conditions and is perpendicular to the baseline in the third point cloud region and the fourth point cloud region respectively to obtain a fourth straight line and a fifth straight line, and use the rectangle enclosed by the baseline, the third straight line, the fourth straight line and the fifth straight line as the rectangular frame of the vehicle to be detected.

[0163] Optionally, the paint film inspection chamber is equipped with multiple three-dimensional radars. The data acquisition module 1201 is specifically used to: control the multiple three-dimensional radars to collect the original point cloud data of the vehicle to be inspected, and perform coordinate transformation on the original point cloud data to obtain the positioning point cloud data of the vehicle to be inspected in the target coordinate system.

[0164] Optionally, the vehicle to be inspected is located on a paint film inspection turntable, and the correction module 1203 is specifically used to: determine the distance offset and angle offset of the vehicle to be inspected based on the original pose and the target pose; if the distance offset is less than or equal to a preset distance threshold and the angle offset is less than or equal to a preset angle threshold, then the pose offset correction of the vehicle to be inspected is determined to be complete; if the distance offset is greater than the preset distance threshold, then a prompt message is displayed, the prompt message being used to prompt the vehicle to be inspected to move to the target distance corresponding to the distance offset; if the angle offset is greater than the preset angle threshold, then the paint film inspection turntable is controlled to rotate to the target angle corresponding to the angle offset.

[0165] Optionally, the paint film detection module 1204 includes: a data acquisition unit for acquiring detection point cloud data and image data of the vehicle to be inspected, wherein the detection point cloud data is three-dimensional radar point cloud data; a modeling unit for performing three-dimensional modeling of the vehicle to be inspected based on the detection point cloud data and the image data to obtain a three-dimensional model of the vehicle to be inspected; a point determination unit for determining the detection point information of the vehicle to be inspected based on the three-dimensional model; and a paint film detection unit for controlling the paint film analyzer to perform paint film detection on the vehicle to be inspected based on the detection point information.

[0166] Optionally, the paint film analyzer is mounted on a robotic arm, which is mounted on an automated guided vehicle (AGV). The point determination unit is specifically used to: determine the target point information of the AGV and at least one detection point information corresponding to the target point information based on the three-dimensional model; the paint film detection unit includes: a movement control subunit, used to control the AGV and the robotic arm to move the paint film analyzer to the detection point corresponding to the detection point information based on the target point information and the detection point information; and a detection control subunit, used to control the paint film analyzer to perform paint film detection on the detection point.

[0167] Optionally, the movement control subunit is specifically used to: control the automated guided vehicle to move to the target point corresponding to the target point information; and after the automated guided vehicle moves to the target point, control the robotic arm to move the paint film analyzer to the detection point corresponding to the detection point information.

[0168] The vehicle paint film inspection device provided in this disclosure can execute the vehicle paint film inspection method provided in any embodiment of this disclosure, and has the corresponding functional modules and beneficial effects for executing the vehicle paint film inspection method. Technical details not described in detail in this embodiment can be found in the vehicle paint film inspection method provided in any embodiment of this disclosure.

[0169] The following is for reference. Figure 13 The diagram illustrates a structural schematic of an electronic device (e.g., a terminal device) 1300 suitable for implementing embodiments of the present disclosure. The terminal device in the embodiments of the present disclosure may include, but is not limited to, mobile terminals such as mobile phones, laptops, digital broadcast receivers, PDAs (personal digital assistants), PADs (tablet computers), PMPs (portable multimedia players), in-vehicle terminals (e.g., in-vehicle navigation terminals), and fixed terminals such as digital TVs and desktop computers. Figure 13 The electronic device shown is merely an example and should not be construed as limiting the functionality and scope of the embodiments disclosed herein.

[0170] like Figure 13As shown, the electronic device 1300 may include a processing unit (e.g., a central processing unit, a graphics processing unit, etc.) 1301, which can perform various appropriate actions and processes according to a program stored in a read-only memory (ROM) 1302 or a program loaded from a storage device 1308 into a random access memory (RAM) 1303. The RAM 1303 also stores various programs and data required for the operation of the electronic device 1300. The processing unit 1301, ROM 1302, and RAM 1303 are interconnected via a bus 1304. An input / output (I / O) interface 1305 is also connected to the bus 1304.

[0171] Typically, the following devices can be connected to I / O interface 1305: input devices 1306 including, for example, touchscreens, touchpads, keyboards, mice, cameras, microphones, accelerometers, gyroscopes, etc.; output devices 1307 including, for example, liquid crystal displays (LCDs), speakers, vibrators, etc.; storage devices 1308 including, for example, magnetic tapes, hard disks, etc.; and communication devices 1309. Communication device 1309 allows electronic device 1300 to communicate wirelessly or wiredly with other devices to exchange data. Although Figure 13 An electronic device 1300 with various devices is shown; however, it should be understood that it is not required to implement or possess all of the devices shown. More or fewer devices may be implemented or possessed alternatively.

[0172] In particular, according to embodiments of this disclosure, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of this disclosure include a computer program product comprising a computer program carried on a non-transitory computer-readable medium, the computer program containing program code for performing the methods shown in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via communication device 1309, or installed from storage device 1308, or installed from ROM 1302. When the computer program is executed by processing device 1301, it performs the functions defined in the methods of embodiments of this disclosure.

[0173] It should be noted that the computer-readable medium described in this disclosure can be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium can be, for example,—but not limited to—an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of a computer-readable storage medium may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage device, magnetic storage device, or any suitable combination thereof. In this disclosure, a computer-readable storage medium can be any tangible medium containing or storing a program that can be used by or in connection with an instruction execution system, apparatus, or device. In this disclosure, a computer-readable signal medium can include a data signal propagated in baseband or as part of a carrier wave, carrying computer-readable program code. Such propagated data signals can take various forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof. A computer-readable signal medium can be any computer-readable medium other than a computer-readable storage medium, which can send, propagate, or transmit a program for use by or in connection with an instruction execution system, apparatus, or device. The program code contained on the computer-readable medium can be transmitted using any suitable medium, including but not limited to: wires, optical fibers, RF (radio frequency), etc., or any suitable combination thereof.

[0174] In some implementations, clients and servers can communicate using any currently known or future-developed network protocol such as HTTP (Hypertext Transfer Protocol) and can interconnect with digital data communication (e.g., communication networks) of any form or medium. Examples of communication networks include local area networks (“LANs”), wide area networks (“WANs”), the Internet (e.g., the Internet of Things), and peer-to-peer networks (e.g., ad hoc peer-to-peer networks), as well as any currently known or future-developed networks.

[0175] The aforementioned computer-readable medium may be included in the aforementioned electronic device; or it may exist independently and not assembled into the electronic device.

[0176] The aforementioned computer-readable medium carries one or more programs, which, when executed by the electronic device, cause the electronic device to: acquire positioning point cloud data of a vehicle to be inspected in a paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data; locate the vehicle to be inspected based on the positioning point cloud data to obtain the original pose of the vehicle to be inspected; perform pose offset correction on the vehicle to be inspected according to the original pose and the target pose; and perform paint film inspection on the vehicle to be inspected after the pose offset correction is completed to obtain the paint film inspection result.

[0177] Computer program code for performing the operations of this disclosure can be written in one or more programming languages ​​or a combination thereof, including but not limited to object-oriented programming languages ​​such as Java, Smalltalk, and C++, as well as conventional procedural programming languages ​​such as the "C" language or similar programming languages. The program code can be executed entirely on the user's computer, partially on the user's computer, as a standalone software package, partially on the user's computer and partially on a remote computer, or entirely on a remote computer or server. In cases involving remote computers, the remote computer can be connected to the user's computer via any type of network—including a local area network (LAN) or a wide area network (WAN)—or can be connected to an external computer (e.g., via the Internet using an Internet service provider).

[0178] The flowcharts and block diagrams in the accompanying drawings illustrate the architecture, functionality, and operation of possible implementations of systems, methods, and computer program products according to various embodiments of this disclosure. In this regard, each block in a flowchart or block diagram may represent a module, segment, or portion of code containing one or more executable instructions for implementing a specified logical function. It should also be noted that in some alternative implementations, the functions indicated in the blocks may occur in a different order than those indicated in the drawings. For example, two consecutively indicated blocks may actually be executed substantially in parallel, and they may sometimes be executed in reverse order, depending on the functions involved. It should also be noted that each block in the block diagrams and / or flowcharts, and combinations of blocks in the block diagrams and / or flowcharts, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or using a combination of dedicated hardware and computer instructions.

[0179] The units described in the embodiments of this disclosure can be implemented in software or hardware. The names of modules do not, in some cases, constitute a limitation on the unit itself.

[0180] The functions described above in this document can be performed, at least in part, by one or more hardware logic components. For example, exemplary types of hardware logic components that can be used, without limitation, include: Field Programmable Gate Arrays (FPGAs), Application-Specific Integrated Circuits (ASICs), Application Standard Products (ASSPs), System-on-Chip (SoCs), Complex Programmable Logic Devices (CPLDs), and so on.

[0181] In the context of this disclosure, a machine-readable medium can be a tangible medium that may contain or store a program for use by or in conjunction with an instruction execution system, apparatus, or device. A machine-readable medium can be a machine-readable signal medium or a machine-readable storage medium. A machine-readable medium can be, but is not limited to, electronic, magnetic, optical, electromagnetic, infrared, or semiconductor systems, apparatus, or devices, or any suitable combination of the foregoing. More specific examples of machine-readable storage media include electrical connections based on one or more wires, portable computer disks, hard disks, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber, portable compact disk read-only memory (CD-ROM), optical storage devices, magnetic storage devices, or any suitable combination of the foregoing.

[0182] According to one or more embodiments of this disclosure, Example 1 provides a method for inspecting the paint film of a vehicle, comprising:

[0183] Acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data;

[0184] The vehicle to be detected is located based on the location point cloud data to obtain the original pose of the vehicle to be detected.

[0185] The vehicle to be detected is subjected to pose offset correction based on the original pose and the target pose.

[0186] After the pose offset correction is completed, the paint film of the vehicle to be tested is inspected to obtain the paint film inspection results.

[0187] According to one or more embodiments of this disclosure, Example 2 describes the method described in Example 1, wherein locating the vehicle to be detected based on the positioning point cloud data to obtain the original pose of the vehicle to be detected includes:

[0188] The location point cloud data is mapped in two dimensions to obtain the target point cloud data of the vehicle to be detected.

[0189] The baseline of the vehicle to be detected is determined based on the target point cloud data;

[0190] A rectangular frame of the vehicle to be inspected is constructed based on the baseline.

[0191] The original pose of the vehicle to be detected is determined based on the rectangular frame.

[0192] According to one or more embodiments of this disclosure, Example 3, based on the method of Example 2, further includes, before performing two-dimensional mapping on the location point cloud data:

[0193] The location point cloud data is filtered using a preset filtering method, which includes at least one of height filtering, distance filtering, and clustering filtering.

[0194] The step of performing two-dimensional mapping on the location point cloud data includes:

[0195] Perform two-dimensional mapping on the filtered location point cloud data.

[0196] According to one or more embodiments of this disclosure, Example 4 describes the method described in Example 2, wherein determining the baseline of the vehicle to be detected based on the target point cloud data includes:

[0197] A first dividing line is drawn based on the target pose. The target point cloud data is divided into a first point cloud region and a second point cloud region by the first dividing line. The first dividing line is parallel to the vehicle body direction corresponding to the target pose and passes through the target feature points of the target point cloud data.

[0198] A baseline of the vehicle to be detected is drawn based on the first point cloud region or the second point cloud region, wherein the associated points of the baseline in the corresponding point cloud region meet preset conditions, and the distance between the associated points of the baseline and the baseline is within a preset distance range.

[0199] According to one or more embodiments of this disclosure, Example 5 describes the method according to Example 4, wherein the preset conditions include a first preset condition and / or a second preset condition, and the step of drawing the baseline of the vehicle to be detected based on the first point cloud region or the second point cloud region includes:

[0200] In the first point cloud region, obtain a first straight line whose proportion of associated points satisfies a first preset condition;

[0201] If the first straight line is obtained, the first straight line is used as the baseline of the vehicle to be detected;

[0202] If the first straight line is not obtained, then a second straight line with a number of associated points satisfying the second preset condition is obtained in the first point cloud region and the second point cloud region respectively, and a second straight line is selected as the baseline of the vehicle to be detected based on the number of associated points of each second straight line.

[0203] According to one or more embodiments of this disclosure, Example 6, based on the method of Example 4, further includes, after drawing the first segmentation line according to the target pose:

[0204] A second dividing line perpendicular to the first dividing line is drawn through the target feature points, and the target point cloud data is divided into a third point cloud region and a fourth point cloud region by the second dividing line.

[0205] The process of constructing the rectangular frame of the vehicle to be detected based on the baseline includes:

[0206] For point cloud regions in the first and second point cloud regions where the baseline is not located, draw a third straight line that satisfies the preset conditions and is parallel to the baseline.

[0207] Draw straight lines that satisfy preset conditions and are perpendicular to the baseline in the third point cloud region and the fourth point cloud region respectively, to obtain the fourth straight line and the fifth straight line.

[0208] The rectangle formed by the baseline, the third straight line, the fourth straight line, and the fifth straight line shall be used as the rectangular frame of the vehicle to be inspected.

[0209] According to one or more embodiments of this disclosure, Example 7 describes the method described in Example 1, wherein multiple three-dimensional radars are deployed inside the paint film inspection chamber, and the acquisition of the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber includes:

[0210] The system controls the multiple 3D radars to acquire the raw point cloud data of the vehicle to be detected, and performs coordinate transformation on the raw point cloud data to obtain the positioning point cloud data of the vehicle to be detected in the target coordinate system.

[0211] According to one or more embodiments of this disclosure, Example 8 describes the method of Example 1, wherein the vehicle to be inspected is located on a paint film inspection turntable, and the step of correcting the pose offset of the vehicle to be inspected based on the original pose and the target pose includes:

[0212] The distance offset and angle offset of the vehicle to be detected are determined based on the original pose and the target pose.

[0213] If the distance offset is less than or equal to a preset distance threshold and the angle offset is less than or equal to a preset angle threshold, then the positional offset correction of the vehicle to be detected is determined to be complete.

[0214] If the distance offset is greater than a preset distance threshold, a prompt message is displayed. The prompt message is used to prompt the vehicle to be detected to move to the target distance corresponding to the distance offset.

[0215] If the angle offset is greater than a preset angle threshold, the paint film detection turntable is controlled to rotate to the target angle corresponding to the angle offset.

[0216] According to one or more embodiments of this disclosure, Example 9 describes the method described in any of Examples 1-8, wherein the paint film inspection of the vehicle to be inspected includes:

[0217] Acquire detection point cloud data and image data of the vehicle to be inspected, wherein the detection point cloud data is three-dimensional radar point cloud data;

[0218] The vehicle to be detected is modeled in three dimensions based on the detection point cloud data and the image data to obtain a three-dimensional model of the vehicle to be detected.

[0219] The detection point information of the vehicle to be detected is determined based on the three-dimensional model;

[0220] Based on the detection point information, the paint film analyzer is controlled to perform paint film detection on the vehicle to be tested.

[0221] According to one or more embodiments of this disclosure, Example 10 describes the method of Example 9, wherein the paint film analyzer is mounted on a robotic arm, the robotic arm is mounted on an automated guided vehicle, and the step of determining the detection point information of the vehicle to be inspected based on the three-dimensional model includes:

[0222] Based on the three-dimensional model, the target location information of the automated guided vehicle and at least one detection point information corresponding to the target location information are determined;

[0223] Based on the detection point information, the paint film analyzer is controlled to perform paint film detection on the vehicle to be inspected, including:

[0224] Based on the target location information and the detection location information, the automated guided vehicle and the robotic arm are controlled to move the paint film analyzer to the detection location corresponding to the detection location information.

[0225] The paint film analyzer is controlled to perform paint film detection at the detection points.

[0226] According to one or more embodiments of this disclosure, Example 11, based on the method described in Example 10, controls the automated guided vehicle and the robotic arm to move the paint film analyzer to the detection point corresponding to the detection point information, according to the target point information and the detection point information, including:

[0227] Control the automated guided vehicle to move to the target point corresponding to the target point information;

[0228] After the automated guided vehicle moves to the target location, the robotic arm is controlled to move the paint film analyzer to the detection location corresponding to the detection location information.

[0229] According to one or more embodiments of this disclosure, Example 12 provides a paint film inspection device for a vehicle, comprising:

[0230] The data acquisition module is used to acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data;

[0231] The positioning module is used to locate the vehicle to be detected based on the positioning point cloud data, so as to obtain the original pose of the vehicle to be detected.

[0232] The correction module is used to correct the pose offset of the vehicle to be detected based on the original pose and the target pose.

[0233] The paint film detection module is used to perform paint film detection on the vehicle under test after the pose offset correction is completed, and obtain the paint film detection result.

[0234] According to one or more embodiments of this disclosure, Example 13 provides an electronic device comprising:

[0235] One or more processors;

[0236] Memory, used to store one or more programs.

[0237] When the one or more programs are executed by the one or more processors, the one or more processors implement the paint film detection method for vehicles as described in any of Examples 1-11.

[0238] According to one or more embodiments of the present disclosure, Example 14 provides a computer-readable storage medium having a computer program stored thereon that, when executed by a processor, implements a paint film detection method for a vehicle as described in any of Examples 1-11.

[0239] According to one or more embodiments of the present disclosure, Example 15 provides a computer program product that, when executed by a computer, causes the computer to implement a paint film detection method for a vehicle as described in any of Examples 1-11.

[0240] The above description is merely a preferred embodiment of this disclosure and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of this disclosure is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the above-described concept. For example, technical solutions formed by substituting the above features with (but not limited to) technical features disclosed in this disclosure that have similar functions.

[0241] Furthermore, while the operations are described in a specific order, this should not be construed as requiring these operations to be performed in the specific order shown or in a sequential order. In certain environments, multitasking and parallel processing may be advantageous. Similarly, while several specific implementation details are included in the above discussion, these should not be construed as limiting the scope of this disclosure. Certain features described in the context of individual embodiments may also be implemented in combination in a single embodiment. Conversely, various features described in the context of a single embodiment may also be implemented individually or in any suitable sub-combination in multiple embodiments.

[0242] Although the subject matter has been described using language specific to structural features and / or methodological logic, it should be understood that the subject matter defined in the appended claims is not necessarily limited to the specific features or actions described above. Rather, the specific features and actions described above are merely illustrative examples of implementing the claims.

Claims

1. A method for inspecting the paint film of a vehicle, characterized in that, include: Acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data; The vehicle to be detected is located based on the location point cloud data to obtain the original pose of the vehicle to be detected. The vehicle to be detected is subjected to pose offset correction based on the original pose and the target pose. After the pose offset correction is completed, the paint film of the vehicle under test is inspected to obtain the paint film inspection results; wherein, The step of locating the vehicle to be detected based on the positioning point cloud data to obtain the original pose of the vehicle to be detected includes: performing two-dimensional mapping on the positioning point cloud data to obtain target point cloud data of the vehicle to be detected; determining the baseline of the vehicle to be detected based on the target point cloud data; constructing a rectangular frame of the vehicle to be detected based on the baseline; and determining the original pose of the vehicle to be detected based on the rectangular frame. The step of determining the baseline of the vehicle to be detected based on the target point cloud data includes: drawing a first dividing line based on the target pose, dividing the target point cloud data into a first point cloud region and a second point cloud region through the first dividing line, wherein the first dividing line is parallel to the vehicle body direction corresponding to the target pose and passes through the target feature points of the target point cloud data; drawing the baseline of the vehicle to be detected based on the first point cloud region or the second point cloud region, wherein the associated points of the baseline in the corresponding point cloud region meet preset conditions, and the distance between the associated points of the baseline and the baseline is within a preset distance range. The preset conditions include a first preset condition and / or a second preset condition. The first preset condition is related to the proportion of associated points in the corresponding point cloud region, and the second preset condition is related to the number of associated points.

2. The method according to claim 1, characterized in that, Before performing two-dimensional mapping on the location point cloud data, the method further includes: The location point cloud data is filtered using a preset filtering method, which includes at least one of height filtering, distance filtering, and clustering filtering. The step of performing two-dimensional mapping on the location point cloud data includes: Perform two-dimensional mapping on the filtered location point cloud data.

3. The method according to claim 2, characterized in that, The step of drawing the baseline of the vehicle to be detected based on the first point cloud region or the second point cloud region includes: In the first point cloud region, obtain a first straight line whose proportion of associated points satisfies a first preset condition; If the first straight line is obtained, the first straight line is used as the baseline of the vehicle to be detected; If the first straight line is not obtained, then a second straight line with a number of associated points satisfying the second preset condition is obtained in the first point cloud region and the second point cloud region respectively, and a second straight line is selected as the baseline of the vehicle to be detected based on the number of associated points of each second straight line.

4. The method according to claim 2, characterized in that, After drawing the first segmentation line based on the target pose, the method further includes: A second dividing line perpendicular to the first dividing line is drawn through the target feature points, and the target point cloud data is divided into a third point cloud region and a fourth point cloud region by the second dividing line. The process of constructing the rectangular frame of the vehicle to be detected based on the baseline includes: For point cloud regions in the first and second point cloud regions where the baseline is not located, draw a third straight line that satisfies the preset conditions and is parallel to the baseline. Draw straight lines that satisfy preset conditions and are perpendicular to the baseline in the third point cloud region and the fourth point cloud region respectively to obtain the fourth straight line and the fifth straight line. The rectangle enclosed by the baseline, the third straight line, the fourth straight line and the fifth straight line is used as the rectangular frame of the vehicle to be detected.

5. The method according to claim 1, characterized in that, The paint film inspection chamber is equipped with multiple 3D radars. The acquisition of the location point cloud data of the vehicle to be inspected within the paint film inspection chamber includes: The system controls the multiple 3D radars to acquire the raw point cloud data of the vehicle to be detected, and performs coordinate transformation on the raw point cloud data to obtain the positioning point cloud data of the vehicle to be detected in the target coordinate system.

6. The method according to claim 1, characterized in that, The vehicle to be inspected is located on a paint film inspection turntable. The step of correcting the pose offset of the vehicle to be inspected based on the original pose and the target pose includes: The distance offset and angle offset of the vehicle to be detected are determined based on the original pose and the target pose. If the distance offset is less than or equal to a preset distance threshold and the angle offset is less than or equal to a preset angle threshold, then the positional offset correction of the vehicle to be detected is determined to be complete. If the distance offset is greater than a preset distance threshold, a prompt message is displayed. The prompt message is used to prompt the vehicle to be detected to move to the target distance corresponding to the distance offset. If the angle offset is greater than a preset angle threshold, the paint film detection turntable is controlled to rotate to the target angle corresponding to the angle offset.

7. The method according to any one of claims 1-6, characterized in that, The process of performing paint film inspection on the vehicle to be inspected includes: Acquire detection point cloud data and image data of the vehicle to be inspected, wherein the detection point cloud data is three-dimensional radar point cloud data; The vehicle to be detected is modeled in three dimensions based on the detected point cloud data and the image data to obtain a three-dimensional model of the vehicle to be detected. The detection point information of the vehicle to be detected is determined based on the three-dimensional model; Based on the detection point information, the paint film analyzer is controlled to perform paint film detection on the vehicle to be tested.

8. The method according to claim 7, characterized in that, The paint film analyzer is mounted on a robotic arm, which is mounted on an automated guided vehicle. Determining the detection point information of the vehicle to be inspected based on the three-dimensional model includes: Based on the three-dimensional model, the target location information of the automated guided vehicle and at least one detection point information corresponding to the target location information are determined; Based on the detection point information, the paint film analyzer is controlled to perform paint film detection on the vehicle to be inspected, including: Based on the target location information and the detection location information, the automated guided vehicle and the robotic arm are controlled to move the paint film analyzer to the detection location corresponding to the detection location information. The paint film analyzer is controlled to perform paint film detection at the detection points.

9. The method according to claim 8, characterized in that, Based on the target location information and the detection location information, controlling the automated guided vehicle and the robotic arm to move the paint film analyzer to the detection location corresponding to the detection location information includes: Control the automated guided vehicle to move to the target point corresponding to the target point information; After the automated guided vehicle moves to the target location, the robotic arm is controlled to move the paint film analyzer to the detection location corresponding to the detection location information.

10. A vehicle paint film inspection device, characterized in that, include: The data acquisition module is used to acquire the positioning point cloud data of the vehicle to be inspected in the paint film inspection chamber, wherein the positioning point cloud data is three-dimensional radar point cloud data; The positioning module is used to locate the vehicle to be detected based on the positioning point cloud data, so as to obtain the original pose of the vehicle to be detected. The correction module is used to correct the pose offset of the vehicle to be detected based on the original pose and the target pose. The paint film detection module is used to perform paint film detection on the vehicle under test after the pose offset correction is completed, and obtain the paint film detection result; wherein, The positioning module is further configured to perform two-dimensional mapping on the positioning point cloud data to obtain target point cloud data of the vehicle to be detected; determine the baseline of the vehicle to be detected based on the target point cloud data; construct a rectangular frame of the vehicle to be detected based on the baseline; and determine the original pose of the vehicle to be detected based on the rectangular frame. The positioning module is further configured to draw a first dividing line based on the target pose, dividing the target point cloud data into a first point cloud region and a second point cloud region through the first dividing line, wherein the first dividing line is parallel to the vehicle body direction corresponding to the target pose and passes through the target feature points of the target point cloud data; and draw a baseline of the vehicle to be detected based on the first point cloud region or the second point cloud region, wherein the associated points of the baseline in the corresponding point cloud region meet preset conditions, and the distance between the associated points of the baseline and the baseline is within a preset distance range. The preset conditions include a first preset condition and / or a second preset condition. The first preset condition is related to the proportion of associated points in the corresponding point cloud region, and the second preset condition is related to the number of associated points.

11. An electronic device, characterized in that, include: At least one processor; as well as A memory communicatively connected to the at least one processor; wherein, The memory stores a computer program that can be executed by the at least one processor, the computer program being executed by the at least one processor to enable the at least one processor to perform the paint film detection method for the vehicle according to any one of claims 1-9.

12. A computer-readable storage medium, characterized in that, The computer-readable storage medium stores computer instructions that, when executed by a processor, implement the paint film detection method for the vehicle as described in any one of claims 1-9.

13. A computer program product, characterized in that, The computer program product includes a computer program that, when executed by a processor, implements the paint film detection method for the vehicle as described in any one of claims 1-9.

Images