Pose detection method and device of auxiliary frame
By using image processing technology to identify the characteristic position line and the reference position line of the subframe, the problem of low efficiency of manual visual confirmation is solved, and efficient and accurate posture detection is achieved, which is suitable for automated production in smart factories.
Patent Information
- Application Number
- CN202510915669.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-07-03
- Publication Date
- 2025-10-03
AI Technical Summary
The existing subframe posture detection method relies on manual visual confirmation, which has low efficiency and accuracy and cannot meet the efficient automation needs of modern smart factories.
By acquiring the current position image of the subframe, performing image processing to obtain the target image, identifying the characteristic position straight line, and comparing it with the reference position straight line, the current position of the subframe is determined.
It achieves efficient and accurate subframe posture detection, improves detection efficiency and accuracy, and meets the automated production requirements of smart factories.
Smart Images
Figure CN120747223A_ABST
Abstract
Description
Technical Field
[0001] The present application relates to the field of image detection technology, and in particular to a method and device for detecting the posture of a subframe. Background Art
[0002] To meet the production requirements of multiple configurations and models, new energy vehicle assembly plants are typically designed for flexible production. The subframe plays a crucial role in the chassis, impacting not only the vehicle's handling and comfort but also its overall safety and stability. Modern smart factories boast a high level of automation and information integration, and subframes and other components are typically assembled on high-precision pallets.
[0003] When assembling the subframe, the reverse side is assembled first, then the reverse side is flipped using a flipping machine. Using an automated flipping machine and the accompanying high-precision pallet for positioning requires strict requirements for the subframe's position before flipping. Currently, the existing method for ensuring the correct subframe position relies on visual confirmation by workers, but this method is relatively inefficient and inaccurate. Summary of the Invention
[0004] In view of the above problems, the present application provides a method and device for detecting the posture of a subframe, which can obtain the current posture of the subframe by comparing the characteristic position line with the reference position line, with high efficiency and accuracy.
[0005] In a first aspect, the present application provides a method for detecting the posture of a subframe, comprising: acquiring a current position image of the subframe at a processing station, and processing the current position image to obtain a target image; obtaining a characteristic position straight line of the subframe based on the target image; wherein the characteristic position straight line is a straight line determined by a first target point and a second target point on the subframe; comparing the characteristic position straight line with a reference position straight line to obtain the current posture of the subframe at the processing station; wherein, when the posture of the subframe is correct, the reference position straight line is parallel to or coincides with the straight line determined by the first target point and the second target point.
[0006] In some specific embodiments, the steps of acquiring a current position image of the subframe at the processing station and processing the current position image to obtain a target image include: acquiring a first current position image of the subframe at the processing station from a first perspective, and a second current position image from a second perspective; and performing image fusion based on the first current position image and the second current position image to obtain the target image.
[0007] In some specific embodiments, the step of performing image fusion based on the first current position image and the second current position image to obtain a target image includes: obtaining feature point sets corresponding to the first current position image and the second current position image respectively; removing erroneous points in the feature point set based on a preset method; and realizing the fusion of the first current position image and the second current position image using a preset method based on the feature point set, and using the obtained fused image as the target image.
[0008] In some specific embodiments, after the step of performing image fusion based on the first current position image and the second current position image to obtain a target image, the following steps are included: if the characteristic position straight line of the subframe cannot be obtained based on the target image, adjusting the first viewing angle and / or the second viewing angle; and executing the step of obtaining a first current position image of the subframe at the first viewing angle on the processing station, and a second current position image at the second viewing angle.
[0009] In some specific embodiments, the reference position line coincides with a line determined by the first target point and the second target point; and the step of obtaining a characteristic position line of the subframe based on the target image includes: obtaining an edge line of a target structure on the subframe based on the target image; wherein the edge line on the subframe is parallel to the line determined by the first target point and the second target point; and using the edge line as the characteristic position line.
[0010] In some specific embodiments, the step of comparing the characteristic position straight line with the reference position straight line to obtain the current posture of the subframe on the processing station includes: comparing the characteristic position straight line with the reference position straight line to obtain the current angle and current distance between the characteristic position straight line and the reference position straight line; comparing the current angle and the current distance with the target angle and the target distance respectively, and obtaining the current posture of the subframe on the processing station based on the comparison results.
[0011] In some specific embodiments, the step of obtaining the edge line of the target structure on the subframe based on the target image includes: processing the target image through a preset edge detection processing method to obtain a binary edge image; processing the binary edge image through a preset transformation processing method to obtain the edge line of the target structure on the subframe.
[0012] In some specific embodiments, the reference position straight line coincides with a straight line determined by center points of a first reference hole and a second reference hole of the subframe; and the step of obtaining a characteristic position straight line of the subframe based on the target image includes: obtaining positions of the first reference hole and the second reference hole on the subframe based on the target image; using the center points of the first reference hole and the second reference hole as first target points and second target points, respectively, and determining the characteristic position straight line of the subframe based on the first target points and the second target points.
[0013] In some specific embodiments, after the step of obtaining the current position image of the subframe at the processing station, the method includes: obtaining a mapping relationship between an intrinsic parameter matrix and an extrinsic parameter matrix of an image acquisition device; and obtaining world coordinate information of the subframe in the current position image based on the current position image and the mapping relationship.
[0014] A second aspect of the present application provides a subframe posture detection device, comprising: a processor; and a memory for storing a computer program, wherein when the computer program is executed by the processor, any of the above-mentioned subframe posture detection methods is implemented.
[0015] The present application has at least one beneficial technical effect: Based on the subframe posture detection method and device provided by the present application, the method includes: obtaining a current position image of the subframe at a processing station, and processing the current position image to obtain a target image; obtaining a characteristic position line of the subframe based on the target image; wherein the characteristic position line is a line defined by a first target point and a second target point on the subframe; comparing the characteristic position line with a reference position line to obtain the current posture of the subframe at the processing station; wherein, when the posture of the subframe is correct, the reference position line is parallel to or coincides with the line defined by the first target point and the second target point. Therefore, by comparing the characteristic position line with the reference position line, the current posture of the subframe can be obtained efficiently and accurately.
[0016] The above description is only an overview of the technical solutions of the embodiments of the present application. In order to more clearly understand the technical means of the embodiments of the present application, they can be implemented in accordance with the contents of the specification. In order to make the above and other purposes, features and advantages of the embodiments of the present application more obvious and easy to understand, the specific implementation methods of the present application are listed below. BRIEF DESCRIPTION OF THE DRAWINGS
[0017] The accompanying drawings are only used to illustrate the embodiments and are not to be considered as limiting the present application. In addition, the same reference symbols are used to represent the same components throughout the drawings. In the drawings: Figure 1 This is a flow chart of an embodiment of a method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 2 1 is a flow chart of another embodiment of a method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 3 1 is a flow chart of another embodiment of the method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 4 1 is a flow chart of another embodiment of the method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 5 1 is a flow chart of another embodiment of the method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 6 It is a schematic diagram of the relationship between the reference position line and the characteristic position line; Figure 7 1 is a flow chart of another embodiment of the method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 8 1 is a flow chart of another embodiment of the method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 9 1 is a flow chart of another embodiment of the method for detecting the position and posture of a vehicle subframe provided by the present application; Figure 10 It is a schematic diagram of the relationship between the reference position line and the characteristic position line; Figure 11 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application. DETAILED DESCRIPTION
[0018] The following describes exemplary embodiments of the present application in more detail with reference to the accompanying drawings. Although the accompanying drawings show exemplary embodiments of the present application, it should be understood that the present application can be implemented in various forms and should not be limited by the embodiments described herein. All other embodiments obtained by persons of ordinary skill in the art based on the embodiments of the present application without inventive effort are within the scope of protection of this application.
[0019] If there are descriptions involving "first", "second", etc. in the embodiments of the present application, the descriptions of "first", "second", etc. are only for descriptive purposes and cannot be understood as indicating or suggesting their relative importance or implicitly indicating the number of the indicated technical features. Therefore, the features limited to "first" and "second" may explicitly or implicitly include at least one of such features. In addition, if the meaning of "and / or" appearing in the full text is to include three parallel schemes, taking "A and / or B" as an example, it includes scheme A, or scheme B, or a scheme in which A and B are satisfied at the same time. In addition, the technical solutions between the various embodiments can be combined with each other, but it must be based on the ability of ordinary technicians in this field to implement. When the combination of technical solutions is mutually contradictory or cannot be implemented, it should be deemed that such a combination of technical solutions does not exist and is not within the scope of protection required by this application.
[0020] The first aspect of the present application provides a method for detecting the posture of a subframe. Figure 1 This is a flow chart of an embodiment of the vehicle subframe posture detection method provided by this application. Figure 1 , the method comprises the following steps: S101: Acquire a current position image of the subframe at the processing station, and process the current position image to obtain a target image.
[0021] Specifically, the subframe processing station can be a station on an assembly line, equipped with a processing device for processing the subframe and a turning device for turning the subframe. After the processing device completes processing on one side of the subframe, the turning device can turn the subframe over so that the processing device can process the other side of the subframe.
[0022] An image acquisition device may be provided at or near the processing station, and is configured to capture an image of the subframe's current position at the processing station. The current position image may include only the subframe, or may include the subframe and surrounding structures. The current position image may reflect the subframe's specific position at the processing station, or relative to the processing station.
[0023] After acquiring the current position image, it is further processed to obtain a target image, also including position information, to obtain more comprehensive information, including the subframe's position. This step does not limit the image processing method; it focuses on image acquisition and image transformation to better obtain the required information in subsequent processes.
[0024] S102: Obtaining a characteristic position straight line of the subframe based on the target image; wherein the characteristic position straight line is a straight line determined by a first target point and a second target point on the subframe.
[0025] After acquiring the target image, this step further processes the target image, and then obtains the characteristic position straight line of the subframe based on the target image.
[0026] It should be understood that the first and second target points on the subframe are predetermined based on actual needs. The first and second target points can be relatively easy to identify points on the subframe, such as points on certain special structures. In this case, the characteristic position line is a line passing through the first and second target points. The target image includes position information, and the subframe's characteristic position line, acquired based on the target image, also includes position information, reflecting the position of the characteristic position line relative to the processing station. In fact, the characteristic position line can effectively reflect the subframe's current specific position at the processing station.
[0027] S103: Comparing the characteristic position line with the reference position line to obtain the current posture of the subframe at the processing station; wherein, when the posture of the subframe is correct, the reference position line is parallel to or coincides with the line determined by the first target point and the second target point.
[0028] Among them, the correct posture of the subframe is set in advance according to the actual situation. When the subframe is in the correct posture, the relative position of the subframe on the processing station is fixed, and the position of the reference position straight line relative to the processing station is also fixed. In addition, when the subframe is in the correct posture, the reference position straight line is parallel to or coincides with the straight line determined by the first target point and the second target point on the subframe. At this time, the straight line determined by the first target point and the second target point when the subframe posture is correct is set as the subframe straight line. The subframe straight line can reflect the position of the subframe relative to the processing station when the subframe posture is correct. Since the reference straight line coincides with or is parallel to the subframe straight line, the reference position straight line can also reflect the position of the subframe relative to the processing station when the subframe posture is correct.
[0029] Combining the above, the characteristic position line can effectively reflect the subframe's current specific position at the processing station, while the reference position line can reflect the subframe's position relative to the processing station when it is in the correct posture. Therefore, by comparing the two, we can determine the positional relationship between the subframe's current position and its correct posture. This positional relationship can reflect the subframe's current posture at the processing station, that is, the subframe's current posture at the processing station.
[0030] In summary, the subframe posture detection method provided in the above embodiment obtains a characteristic position line through image acquisition and processing, and compares the characteristic position line with the reference position line to obtain the current posture of the subframe, which is efficient and accurate.
[0031] Figure 2 It is a flow chart of another embodiment of the vehicle subframe posture detection method provided by the present application.
[0032] Combine Figure 2 In some specific embodiments, the step of obtaining the current position image of the subframe at the processing station and processing the current position image to obtain the target image, i.e., the above-mentioned step S101, includes: S201: Acquire a first current position image of the sub-frame at a first viewing angle on a processing station, and a second current position image at a second viewing angle.
[0033] Among them, the first perspective and the second perspective can be set in advance according to actual needs. The first perspective and the second perspective can be perspectives with a large difference, so that the first current position image and the second current position image obtained under the first perspective and the second perspective respectively can more comprehensively reflect the position of the entire subframe on the processing station.
[0034] In the specific implementation process, different image acquisition devices can be set to acquire images at different perspectives, and different image acquisition devices can be set at different positions of the processing station, thereby better meeting the acquisition of images at different perspectives.
[0035] S202: Perform image fusion based on the first current position image and the second current position image to obtain a target image.
[0036] It should be understood that the first and second current-position images may focus on reflecting a particular portion of the subframe structure, failing to adequately reflect other portions or failing to reflect other portions at all. Therefore, in this step, image fusion is performed based on the first and second current-position images to generate a target image. This target image integrates information from images at various viewing angles, thereby enabling the target image to comprehensively and clearly reflect information from all portions of the subframe.
[0037] Of course, the above embodiment is merely an example of how to obtain a target image based on the fusion of images from two perspectives. In other embodiments, images from three or more perspectives may be obtained and then fused to obtain the target image.
[0038] Figure 3 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application.
[0039] Combine Figure 3 In some specific embodiments, the step of performing image fusion based on the first current position image and the second current position image to obtain the target image, i.e., the above-mentioned step S202, includes: S301: Acquire feature point sets corresponding to the first current position image and the second current position image respectively.
[0040] Both the first current position image and the second current position image have a large number of points. Feature points can be points that can well reflect certain features of the images. A collection of feature points constitutes a feature point set. The type of feature point can be pre-set based on actual conditions. For example, a feature point can be a point on the edge structure of a subframe. In this case, the feature point can well reflect the edge structure of the subframe, specifically the edge line formed by the edge structure on the subframe.
[0041] In some application scenarios, feature points can be points on the edge structure of the subframe. In this case, the feature point set is the edge feature point set. In this case, the ORB feature point set in the target image can be extracted based on the improved ORB-SLAM calculation. ORB-SLAM is a feature point-based visual simultaneous localization and mapping system. In short, it is a powerful algorithm framework.
[0042] At this point, the process of calculating the ORB feature point set can be as follows: Among them, xi and yi represent the pixel coordinates of the feature points in the image, θi represents the main direction angle of the feature point, di represents the binary BRIEF descriptor, and N represents the total number of feature points extracted from a single frame image.
[0043] S302: Remove erroneous points in the feature point set based on a preset method.
[0044] It should be understood that some feature points in the feature point set determined by the above method may be erroneous. For example, when the feature point set is an edge feature point set, some points may not be edge feature points. In this case, these non-edge feature points need to be removed from the edge feature point set. The preset method can be set in advance based on actual needs, and there are multiple preset methods.
[0045] It should be understood that different position images correspond to different feature point sets. In this case, the first position image and the second current position image correspond to the first feature point set and the second feature point set, respectively. In this step, the error points in the first feature point set and the error points in the second feature point set are removed respectively.
[0046] In some application scenarios, bidirectional optical flow constraints can be used to remove erroneous points in the feature point set. The calculation process can be as follows: Where pk represents the coordinates of the kth feature point in the reference image, qk represents the corresponding feature in the target image, W(qk,vk) represents the nonlinear transformation function of the optical flow field, vk represents the optical flow vector estimated by the Lucas-Kanade algorithm, and ϵ represents the matching tolerance threshold.
[0047] S303: Based on the feature point set, a preset method is used to fuse the first current position image and the second current position image, and the obtained fused image is used as the target image.
[0048] After obtaining the feature point set corresponding to the image, a preset method is further used based on the feature point set to fuse the first current position image with the second current position image. The preset method can also be set in advance based on actual needs. In addition to being based on the feature point set, the fusion process can also use some other image information, which is not specifically limited here.
[0049] In combination with the above content, if the feature point set is an edge feature point set, then in the process of image fusion based on the feature point set, the edge line of the fused image can be obtained, and then an accurate fused image can be obtained as the target image.
[0050] In some application scenarios, a multi-band wavelet fusion algorithm can be used to achieve the fusion of the first current position image and the second current position image. The specific fusion process calculation can be as follows: Where Ifused(x, y) represents the grayscale value of the fused panoramic image at the pixel point (x, y), ∑k represents the accumulation operation of weighted fusion of the source images collected from three different perspectives, wk represents the weight coefficient of the k-th perspective in the fusion process, and W(•) and W-1(•) represent two-dimensional discrete wavelet transform.
[0051] Figure 4 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application.
[0052] Combine Figure 4 In combination with the above content, in some specific embodiments, after the step of performing image fusion based on the first current position image and the second current position image to obtain the target image, the method includes: S401: If the characteristic position straight line of the sub-frame cannot be obtained based on the target image, adjust the first viewing angle and / or the second viewing angle.
[0053] In light of the above, after obtaining the target image, the subframe's characteristic position line is further determined based on the target image. If the subframe's characteristic position line cannot be determined based on the target image, it indicates that there is a problem with the target image, potentially resulting in an incomplete image or unclear image features. In this case, to obtain a better target image, the first current position image and / or the second current position image must be modified, namely by adjusting the first and / or second viewing angles.
[0054] It should be understood that the viewing angle can be adjusted by adjusting the position of the image acquisition device or by adjusting the camera angle of the image acquisition device, and no specific limitation is given here.
[0055] S402: Execute the step of acquiring a first current position image of the sub-frame at a first viewing angle on a processing station, and a second current position image at a second viewing angle.
[0056] After adjusting the first viewing angle and / or the second viewing angle, continue to execute the step of obtaining the first current position image of the sub-frame at the processing station at the first viewing angle and the second current position image at the second viewing angle and subsequent steps, that is, continue to execute the above-mentioned step S201 and subsequent steps.
[0057] Figure 5 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application.
[0058] In some specific embodiments, when the sub-frame is in correct position, the reference position straight line is parallel to the straight line determined by the first target point and the second target point. Figure 5 The step of obtaining the characteristic position straight line of the subframe based on the target image, i.e., the above-mentioned step S102, includes: S501: Acquire an edge line of a target structure on the subframe based on a target image; wherein the edge line on the subframe coincides with a straight line determined by a first target point and a second target point.
[0059] At this point, there are some structures on the subframe whose edge lines are parallel to the straight line determined by the first and second target points, that is, the edge lines pass through the first and second target points. At this point, we can select some structures from these structures as target structures.
[0060] S502: Use the edge line as a feature position line.
[0061] As can be seen from the above, the edge line now coincides with the straight line determined by the first target point and the second target point. At this point, the edge line can be used as a feature position straight line.
[0062] Figure 6 It is a schematic diagram of the setting relationship between the reference position line and the characteristic position line.
[0063] Combine Figure 6 , the first target point on the subframe is A, and the second target point is B, then the straight line L2 passing through AB is the characteristic position straight line L2, which is also the edge line of the target structure in this embodiment. In this embodiment, when the subframe is in the correct position, the reference position straight line is parallel to the straight line determined by the first target point and the second target point. Therefore, combined with Figure 6 If the subframe posture is correct, the reference position straight line L1 is parallel to the characteristic position straight line L2 determined by the first target point A and the second target point B.
[0064] Continue to combine Figure 6 In this embodiment, the reference position straight line L1 can be determined by the center points of the two reference holes on the subframe.
[0065] Figure 7 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application.
[0066] Combine Figure 7 In some specific embodiments, the step of comparing the characteristic position line with the reference position line to obtain the current position of the subframe at the processing station, i.e., the above-mentioned step S103, includes: S601: Compare the characteristic position straight line with the reference position straight line to obtain the current angle and the current distance between the characteristic position straight line and the reference position straight line.
[0067] In this embodiment, the purpose of comparing the positions of the characteristic position straight line and the reference position straight line is to obtain the current angle and the current distance between the characteristic position straight line and the reference position straight line.
[0068] S602: Compare the current angle and the current distance with the target angle and the target distance respectively, and obtain the current posture of the subframe on the processing station according to the comparison results.
[0069] At this time, the current angle is compared with the target angle, and the current distance is compared with the target distance, thereby obtaining two comparison results, and the current position of the subframe on the processing station is obtained through these two comparison results.
[0070] In some application scenarios, when the current angle is greater than the target angle, or the current distance is greater than the target distance, the subframe's current position at the processing station can be considered inconsistent with its correct position. When the current position is inconsistent with the correct position, the relevant equipment can be controlled to stop working and then adjusted to align the current position of the subframe with the correct position, at which point the relevant equipment can be controlled to continue working.
[0071] Figure 8 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application.
[0072] Combine Figure 8 In some specific embodiments, the step of acquiring the edge line of the target structure on the subframe based on the target image, i.e., the above-mentioned step S501, includes: S701: Processing the target image using a preset edge detection processing method to obtain a binary edge image.
[0073] In combination with the above content, at this time the reference position line is parallel to the line determined by the first target point and the second target point, and the edge line of the target structure will be used as the reference position line. This embodiment further restricts the acquisition of the edge line of the target structure.
[0074] The preset edge detection processing method may be a preset processing method that can process a binary edge image. In some application scenarios, the preset edge detection processing method may be Canny edge detection, and the calculation process may be as follows: Gradient Strength = , gradient direction = Where σ represents the standard deviation of the Gaussian kernel, x and y represent the coordinates within the kernel, and Thigh and Tlow represent the thresholds.
[0075] S702: Processing the binary edge image using a preset transformation processing method to obtain edge lines of the target structure on the subframe.
[0076] After obtaining the binarized edge image, the binarized edge image is further processed using a preset transformation processing method to obtain the edge line of the target structure on the subframe. The purpose of the preset transformation processing method is to obtain the edge line of the target structure on the subframe. Therefore, any processing method that can achieve this purpose can be used as the preset transformation processing method.
[0077] In some application scenarios, the preset transformation processing method may be Hough transform, and its calculation process is as follows: Detect straight line, accumulator: (ρ,θ)≥Though Among them, ρ represents the normal distance from the straight line to the origin, θ represents the angle between the normal and the x-axis, x and y represent the coordinates of the edge point in the image space, and Through represents the threshold.
[0078] Figure 9 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided in the present application.
[0079] Combine Figure 9 The step of obtaining the characteristic position straight line of the subframe based on the target image, i.e., the above-mentioned step S102, includes: S801: Obtaining positions of a first reference hole and a second reference hole on the subframe based on the target image.
[0080] Figure 10 It is a schematic diagram of the setting relationship between the reference position line and the characteristic position line.
[0081] In this embodiment, when the sub-frame is in the correct position, the reference position straight line coincides with the straight line determined by the center points of the first reference hole and the second reference hole of the sub-frame. Figure 10 If the subframe is in the correct position, the reference position straight line L3 coincides with the straight line determined by the center points of the first reference hole and the second reference hole.
[0082] Wherein, obtaining the positions of the first reference hole and the second reference hole on the sub-frame based on the target image can be achieved through a preset target detection method, which is not specifically limited here.
[0083] S802: The center points of the first reference hole and the second reference hole are used as the first target point and the second target point respectively, and a characteristic position straight line of the subframe is determined based on the first target point and the second target point.
[0084] After obtaining the positions of the first reference hole and the second reference hole, the center points of the first reference hole and the second reference hole are further used as the first target point and the second target point respectively, and the characteristic position straight line of the subframe is determined based on the first target point and the second target point. Figure 10 , if the posture of the subframe is correct at this time, the characteristic position straight line coincides with the reference position straight line.
[0085] Figure 11 This is a flow chart of another embodiment of the vehicle subframe posture detection method provided by the present application. In some specific embodiments, after the step of obtaining the current position image of the subframe at the processing station, the method includes: S901: Acquire a mapping relationship between an intrinsic parameter matrix and an extrinsic parameter matrix of an image acquisition device.
[0086] It should be understood that image acquisition devices, such as cameras, have intrinsic matrix (Intrinsic Matrix, usually denoted as K) and extrinsic matrix (Extrinsic Matrix, usually expressed as [R | t]), which are the core concepts in computer vision, photogrammetry and robotics to describe camera imaging geometry. Together, they constitute the projection model.
[0087] The intrinsic parameter matrix describes the inherent optical and geometric properties of the camera itself and enables the conversion between three-dimensional points in the image acquisition device's coordinate system (with the optical center of the image acquisition device as the origin) and two-dimensional points on the image plane coordinate system (pixel coordinate system). The extrinsic parameter matrix describes the camera's position and orientation (posture) in the world coordinate system and enables the conversion between three-dimensional points in the world coordinate system and points in the image acquisition device's coordinate system.
[0088] S902: Based on the current position image and the mapping relationship, obtain world coordinate information of the subframe in the current position image.
[0089] Therefore, the world coordinate information of the subframe in the current position image can be obtained through the position information in the current image and the above-mentioned mapping relationship. The world coordinate information is included in the current position image.
[0090] It should be understood that coordinate systems include the world coordinate system and other coordinate systems. The world coordinate system is an absolute, global reference frame. It is the unified coordinate system based on which the entire scene or environment is defined. The world coordinate system corresponds to world coordinate information. Other coordinate systems are local coordinate systems defined relative to a specific reference object (such as the object itself, the camera, a joint, etc.). They exist to facilitate the description and manipulation of local relationships.
[0091] A second aspect of the present application provides a subframe posture detection device, comprising: a processor; and a memory for storing a computer program, wherein when the computer program is executed by the processor, the subframe posture detection method in any of the above embodiments is implemented.
[0092] The processor may be a central processing unit (CPU), and the memory may be a read-only memory (ROM). The CPU can perform various appropriate actions and processes, such as executing the method described in the above embodiment, based on programs stored in the ROM or programs loaded from the storage unit into the random access memory (RAM). The RAM also stores various programs and data required for system operation.
[0093] In summary, the method and device for detecting the posture of a subframe provided in this application include: acquiring an image of the subframe's current position at a processing station, processing the current position image to obtain a target image; obtaining a characteristic position line of the subframe based on the target image; wherein the characteristic position line is a line defined by a first target point and a second target point on the subframe; and comparing the characteristic position line with a reference position line to obtain the subframe's current posture at the processing station; wherein, when the subframe's posture is correct, the reference position line is parallel to or coincides with the line defined by the first target point and the second target point. Therefore, by comparing the characteristic position line with the reference position line, the current posture of the subframe can be obtained efficiently and accurately.
[0094] The above content is only a preferred exemplary embodiment of the present application and is not intended to limit the implementation scheme of the present application. Ordinary technicians in this field can easily make corresponding changes or modifications based on the main ideas and spirit of the present application. Therefore, the scope of protection of the present application shall be based on the scope of protection required by the claims.
Claims
1. A method for detecting the posture of a subframe, characterized in that: include: Acquiring a current position image of the subframe at the processing station, and processing the current position image to obtain a target image; Obtaining a characteristic position straight line of the subframe based on the target image; wherein the characteristic position straight line is a straight line determined by a first target point and a second target point on the subframe; The characteristic position straight line is compared with the reference position straight line to obtain the current posture of the subframe on the processing station; wherein, when the posture of the subframe is correct, the reference position straight line is parallel to or coincides with the straight line determined by the first target point and the second target point.
2. The method for detecting the position and posture of a subframe according to claim 1, wherein: The steps of acquiring a current position image of the subframe at the processing station and processing the current position image to obtain a target image include: Acquire a first current position image of the subframe at a first viewing angle on a processing station, and a second current position image at a second viewing angle; Image fusion is performed based on the first current position image and the second current position image to obtain a target image.
3. The method for detecting the position and posture of a subframe according to claim 2, wherein: The step of performing image fusion based on the first current position image and the second current position image to obtain a target image includes: Acquire feature point sets corresponding to the first current position image and the second current position image respectively; removing erroneous points in the feature point set based on a preset method; The first current position image and the second current position image are fused in a preset manner based on the feature point set, and the obtained fused image is used as the target image.
4. The method for detecting the position and posture of a subframe according to claim 2, wherein: After the step of performing image fusion based on the first current position image and the second current position image to obtain a target image, the method includes: If the characteristic position straight line of the subframe cannot be obtained based on the target image, adjusting the first viewing angle and / or the second viewing angle; The steps of acquiring a first current position image of the sub-frame at a first viewing angle on a processing station and a second current position image at a second viewing angle are performed.
5. The method for detecting the position and posture of a subframe according to claim 1, wherein: When the subframe is in a correct position, the reference position straight line is parallel to a straight line determined by the first target point and the second target point; The step of obtaining the characteristic position straight line of the subframe based on the target image includes: Acquire an edge line of a target structure on the subframe based on the target image; wherein the edge line on the subframe coincides with a straight line determined by the first target point and the second target point; The edge line is used as a feature position line.
6. The method for detecting the position and posture of a subframe according to claim 5, wherein: The step of comparing the characteristic position straight line with the reference position straight line to obtain the current position of the subframe on the processing station includes: Comparing the characteristic position straight line with the reference position straight line to obtain a current angle and a current distance between the characteristic position straight line and the reference position straight line; The current angle and the current distance are compared with the target angle and the target distance respectively, and the current posture of the subframe on the processing station is obtained according to the comparison results.
7. The method for detecting the position and posture of a subframe according to claim 5, wherein: The step of acquiring an edge line of a target structure on the subframe based on the target image includes: Processing the target image using a preset edge detection processing method to obtain a binary edge image; The binarized edge image is processed by a preset transformation processing method to obtain an edge line of the target structure on the subframe.
8. The method for detecting the position and posture of a subframe according to claim 1, wherein: When the subframe is in a correct position, the reference position straight line coincides with a straight line defined by center points of the first reference hole and the second reference hole of the subframe; The step of obtaining the characteristic position straight line of the subframe based on the target image includes: Obtaining positions of a first reference hole and a second reference hole on the subframe based on the target image; The center points of the first reference hole and the second reference hole are used as the first target point and the second target point respectively, and a characteristic position straight line of the subframe is determined based on the first target point and the second target point.
9. The method for detecting the position and posture of a subframe according to claim 1, wherein: After the step of obtaining the current position image of the subframe at the processing station, the following steps are included: Acquire a mapping relationship between an intrinsic parameter matrix and an extrinsic parameter matrix of an image acquisition device; Based on the current position image and the mapping relationship, world coordinate information of the subframe in the current position image is obtained.
10. A device for detecting the posture of a subframe, characterized in that: include: processor; A memory for storing a computer program, wherein when the computer program is executed by the processor, the method for detecting the posture of the subframe according to any one of claims 1 to 9 is implemented.