A multi-view vision ranging method for power battery module size

Abstract

The application relates to the field of battery production detection, and discloses a multi-view vision ranging method for the size of a power battery module, which comprises the following steps: placing a standard piece at a fixed position for measuring the size of a battery module on a logistics line, and placing a calibration plate on the surface of the standard piece and at a position corresponding to a circular hole in the battery module; placing a camera corresponding to the calibration plate at a position corresponding to the circular hole in the battery module and at the same set height as the logistics line; photographing the calibration plate by using the camera at each position to obtain a transformation matrix corresponding to each camera; removing the calibration plate, photographing the standard piece by using the camera at each position to correct the transformation matrix and obtain a target transformation matrix in the same coordinate system; photographing the battery module by using the camera at each position when the battery module is produced on the logistics line, extracting the image coordinates of the corner points of the battery module, and combining the target transformation matrix to obtain the size of the battery module. The size of the battery module obtained in this way is high in accuracy and low in cost.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of battery production detection, in particular to a multi-view vision ranging method for the size of a power battery module. BACKGROUND

[0002] In the field of industrial production of power batteries, after the negative and positive anode single-piece battery cells are paired and formed, they need to flow into the subsequent process for packaging operation. In order to ensure good contact of the electrodes between the battery modules and the convenience of welding, it is necessary to ensure that the sizes of the battery modules are consistent. At the same time, defective bare battery cells deformed and expanded during the winding and hot pressing processes can also be detected again according to the measurement results of the module size.

[0003] At present, the measurement of the size of a battery module mainly involves taking a picture of the entire battery module through a camera first, and then processing the image to calculate the size of the module. However, power batteries have the characteristics of strong power storage capacity and large module size. In general, when taking a picture covering the entire battery module, a long-focus high-definition camera with a long far field of view is needed, which not only has a high cost, but also has a small amount of space in an automatic production machine tool. The far field of view camera needs to be installed outside the machine tool, which will be blocked by some machine parts or reflect light during shooting, making it difficult for subsequent image processing. SUMMARY

[0004] Therefore, the purpose of the present application is to provide a multi-view vision ranging method for the size of a power battery module, which can accurately obtain the size of the battery module at a low cost. The specific scheme is as follows:

[0005] A multi-view vision ranging method for the size of a power battery module, comprising:

[0006] placing a standard part at a fixed position for measuring the size of a battery module on a logistics line, and placing a calibration plate on the surface of the standard part and at a position corresponding to a circular hole in the battery module;

[0007] placing a camera corresponding to the calibration plate at a position corresponding to the circular hole in the battery module at the same height as the logistics line;

[0008] using the camera at each position to take a picture of the calibration plate to obtain a transformation matrix corresponding to each camera;

[0009] removing the calibration plate and using the camera at each position to take a picture of the standard part to correct the transformation matrix to obtain a target transformation matrix in the same coordinate system;

[0010] when the logistics line is in production, using the camera at each position to take a picture of the battery module, extracting the image coordinates of the corner points of the battery module, and combining the target transformation matrix to obtain the size of the battery module.

[0011] Preferably, in the multi-view vision ranging method for the size of the power battery module provided in the embodiment of the present application, the transformation matrix corresponding to each camera is obtained, comprising:

[0012] The image coordinates of the centers of the circles in the photographed calibration board image are extracted to form a center matrix;

[0013] The physical matrix formed by the physical coordinates of the centers of the circles on each calibration board is subjected to a transmission transformation with the center matrix to obtain the transformation matrix corresponding to each camera.

[0014] Preferably, in the multi-view vision ranging method for the size of the power battery module provided in the embodiment of the present application, the extraction of the image coordinates of the centers of the circles in the photographed calibration board image to form a center matrix comprises:

[0015] For the photographed calibration board image, a ROI rectangular region containing 3x3 circles is selected;

[0016] The calculation of the circle center image coordinates is performed in the ROI rectangular region;

[0017] The calculated 3x3 circle center image coordinates are sorted in a set order to form a center matrix.

[0018] Preferably, in the multi-view vision ranging method for the size of the power battery module provided in the embodiment of the present application, the sorting of the calculated image coordinates of the centers of the 3x3 circles comprises:

[0019] According to the center point and the rotation angle of the ROI rectangular region, the ROI rectangular region is subjected to a center rotation, and the 3x3 circle center image coordinates are also subjected to a rotation operation;

[0020] The ROI rectangular region after rotation is divided into 3x3 small rectangles according to the rectangular vertex coordinates after rotation, and the 3x3 small rectangles are sequentially traversed in a Z-type manner, and when it is found that a certain circle center falls within the nth small rectangle, the circle center image coordinates are arranged in the nth position in the center matrix in a Z-type manner to obtain the 3x3 circle center image coordinates sorted in a Z-type manner; n is a positive integer.

[0021] Preferably, in the multi-view vision ranging method for the size of the power battery module provided in the embodiment of the present application, the transmission transformation of the physical matrix formed by the physical coordinates of the centers of the circles on each calibration board with the center matrix to obtain the transformation matrix corresponding to each camera comprises:

[0022] The offset angle of each calibration board with the horizontal direction is calculated according to the 3x3 circle center image coordinates;

[0023] According to the offset angle, 3*3 modified circle center physical coordinates of a physical coordinate system in which each of the calibration plates is located after being rotated to be parallel to an image coordinate system are obtained;

[0024] The 3*3 circle center image coordinates and the corresponding modified circle center physical coordinates are subjected to a perspective transformation to obtain a transformation matrix corresponding to each camera.

[0025] Preferably, in the multi-view vision distance measurement method for the size of the power battery module provided in the embodiment of the application, the following formula is used to obtain the 3*3 modified circle center physical coordinates of the physical coordinate system in which each of the calibration plates is located after being rotated to be parallel to the image coordinate system:

[0026] Xep=Xp*cosα-Yp*sinα

[0027] Yep=Xp*sinα+Yp*cosα

[0028] Wherein, α represents the offset angle; (Xp, Yp) represents the physical coordinates of the 3*3 circle centers; (Xep, Yep) represents the 3*3 modified circle center physical coordinates.

[0029] Preferably, in the multi-view vision distance measurement method for the size of the power battery module provided in the embodiment of the application, the following formula is used to obtain the transformation matrix corresponding to each camera:

[0030]

[0031] Wherein, is the transformation matrix, A, B, D, E represent the rigid body transformation quantities such as rotation, C, F represent the horizontal and vertical translation quantities respectively, and (Xs, Ys) represents the 3*3 circle center image coordinates.

[0032] Preferably, in the multi-view vision distance measurement method for the size of the power battery module provided in the embodiment of the application, the transformation matrix is modified to obtain a target transformation matrix in the same coordinate system, which comprises:

[0033] For the circle center of the circle hole image of different orientations on the standard part, the circle center coordinates of the corresponding circle hole image are extracted;

[0034] The circle center physical coordinates of the circle holes of different orientations are calculated by using the transformation matrix and the circle center coordinates of the circle hole image.

[0035] The starting point physical coordinates are translated and calculated with the circle center physical coordinates of the circle holes of different orientations to modify the transformation matrix, so as to obtain the target transformation matrix corresponding to the cameras of different orientations in the same coordinate system.

[0036] Preferably, in the multi-view vision ranging method for the size of the power battery module provided in the embodiment of the present application, the preset starting point physical coordinate is translated with the center physical coordinates of the circular holes in different directions to perform translation calculation, so as to correct the transformation matrix, and obtain the target transformation matrix corresponding to the cameras in different directions in the same coordinate system, including:

[0037] The center physical coordinates of the circular holes in one direction are taken as the origin of the physical coordinate system, the horizontal and vertical translation variables of the transformation matrix corresponding to the camera in this direction are calibrated, and the target transformation matrix corresponding to the camera in this direction is obtained.

[0038] According to the center physical coordinates of the circular holes in other directions, the distances from the center of the circular holes in other directions to the origin of the physical coordinate system are calculated.

[0039] According to the calculated distances from the center of the circular holes in other directions to the origin of the physical coordinate system, the target transformation matrix corresponding to the camera in other directions is obtained.

[0040] Preferably, in the multi-view vision ranging method for the size of the power battery module provided in the embodiment of the present application, the image coordinates of the corner points of the battery module are extracted and combined with the target transformation matrix to obtain the size of the battery module, including:

[0041] Edge extraction is performed on the photographed battery module image.

[0042] The image coordinates of the corner points in different directions are calculated according to the extracted edges.

[0043] The image coordinates of the corner points in each direction are combined with the corresponding target transformation matrix to calculate the physical coordinates of the corner points in each direction.

[0044] The size of the battery module is obtained according to the physical coordinates of the corner points in each direction.

[0045] As can be seen from the above technical solutions, the multi-view vision ranging method for the size of the power battery module provided by the present application includes: placing a standard part at a fixed position for measuring the size of the battery module on a logistics line, and placing a calibration plate on the surface of the standard part and in the direction corresponding to the circular hole in the battery module; placing a camera corresponding to the calibration plate at the same set height as the logistics line and in the direction corresponding to the circular hole in the battery module; photographing the calibration plate by using the camera in each direction to obtain the transformation matrix corresponding to each camera; removing the calibration plate and photographing the standard part by using the camera in each direction to correct the transformation matrix and obtain the target transformation matrix in the same coordinate system; when the battery module is produced online on the logistics line, the battery module is photographed by using the camera in each direction, the image coordinates of the corner points of the battery module are extracted and combined with the target transformation matrix to obtain the size of the battery module.

[0046] The multi-view vision distance measuring method for the power battery module size provided by the application can place a standard part at a fixed position for measuring the battery module size on a logistics line, place calibration plates in different directions on the surface of the standard part, use cameras in different directions to take corresponding pictures, map the transformation matrix corresponding to different cameras to the same coordinate system according to the pictures, accurately obtain the size of the battery module according to the transformation matrix in the same coordinate system and the image coordinates of the battery module corner points, and then judge whether the size of the battery module is qualified within the allowed range of specifications, and the cost is low. BRIEF DESCRIPTION OF DRAWINGS

[0047] In order to more clearly illustrate the technical solutions in the embodiments of the application or the related art, the following will briefly introduce the drawings needed to be used in the embodiments or the related art description. Obviously, the drawings in the following description only belong to the embodiments of the application, and for those skilled in the art, other drawings can also be obtained without creative labor on the basis of the provided drawings.

[0048] Figure 1 The flow chart of the multi-view vision distance measuring method for the power battery module size provided by the embodiment of the application;

[0049] Figure 2 The structural schematic diagram of the battery module provided by the embodiment of the application;

[0050] Figure 3 The structural schematic diagram of the standard part and the calibration plate in different directions provided by the embodiment of the application;

[0051] Figure 4 The structural schematic diagram of the camera in different directions provided by the embodiment of the application;

[0052] Figures 5 to 8 The calibration plate image taken in four directions respectively provided by the embodiment of the application;

[0053] Figure 9 The ROI rectangular region schematic diagram containing 3*3 circles provided by the embodiment of the application;

[0054] Figure 10 The schematic diagram of 3*3 circles arranged in Z type provided by the embodiment of the application;

[0055] Figure 11 The schematic diagram of 3*3 circles arranged in N type provided by the embodiment of the application;

[0056] Figure 12 The schematic diagram of the ROI rectangular four vertex coordinates before rotation provided by the embodiment of the application;

[0057] Figure 13A schematic diagram showing the coordinates of the four vertices of the rotated ROI rectangle provided in an embodiment of the present invention;

[0058] Figure 14 This is a schematic diagram illustrating the division of the rotated ROI rectangle into 3×3 smaller rectangles, as provided in an embodiment of the present invention.

[0059] Figure 15 A schematic diagram of the coordinates of the center images of 3×3 circles arranged in a Z-shape, provided in an embodiment of the present invention;

[0060] Figure 16 This is a schematic diagram showing the offset angle between the calibration plate and the horizontal direction provided in an embodiment of the present invention.

[0061] Figure 17 An image of a circular hole taken from the left front position, provided in an embodiment of the present invention;

[0062] Figure 18 This is a schematic diagram of the corner coordinates of a battery module provided in an embodiment of the present invention. Detailed Implementation

[0063] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0064] This invention provides a multi-view visual ranging method for measuring the size of a power battery module, such as... Figure 1 As shown, it includes the following steps:

[0065] S101. Place a standard part at a fixed position on the logistics line where the size of the battery module is measured, and place a calibration plate on the surface of the standard part in the position corresponding to the circular hole in the battery module.

[0066] like Figure 2 As shown, there is a noticeable round hole on each of the four positions of the battery module: front left, rear left, front right, and rear right. When executing step S101, as... Figure 3 As shown, at a fixed position on the horizontal logistics line where the size of the battery module is measured, a standard part similar in size to the battery module (i.e., the standard part is the battery module standard part) is placed. Calibration plates with negligible thickness are placed on the standard part at the corresponding four positions: left front, left rear, right front, and right rear. Specifically, these include left front calibration plate 11, left rear calibration plate 12, right front calibration plate 13, and right rear calibration plate 14.

[0067] S102. A camera corresponding to the calibration plate is placed at the same set height as the logistics line and in the position corresponding to the round hole in the battery module.

[0068] like Figure 4 As shown, four small-field-of-view area array cameras can be fixed at a set height A at the same distance from the logistics line: left front, left rear, right front, and right rear. Specifically, these include left front camera 21, left rear camera 22, right front camera 23, and right rear camera 24.

[0069] It should be noted that the order of steps S101 and S102 is not important. The standard parts, calibration plate and camera can be placed in sequence, or the camera, standard parts and calibration plate can be placed in sequence.

[0070] S103. Use the camera in each direction to take pictures of the calibration board to obtain the transformation matrix corresponding to each camera.

[0071] S104. Remove the calibration plate and use cameras in each direction to photograph the standard part in order to correct the transformation matrix and obtain the target transformation matrix in the same coordinate system.

[0072] Specifically, images are captured after the calibration plate is removed from the standard component. This operation is performed on all four cameras. Since the center distance of each hole on the standard component is known, the transformation matrices corresponding to the four cameras can be mapped to the same coordinate system.

[0073] S105. During online production on the logistics line, cameras from each direction are used to photograph the battery module, and the image coordinates of the corner points of the battery module are extracted and combined with the target transformation matrix to obtain the size of the battery module.

[0074] Specifically, when the logistics line is in normal production, the coordinates of the corner points of the battery module in the image coordinate system can be obtained by using the photographed battery module image. Multiplying the coordinates by the target transformation matrix gives the physical coordinates of the corner points. Based on the obtained physical coordinates of the corner points in each direction, the length and width dimensions of the module can be calculated.

[0075] It should be noted that in the field of computer graphics, the physical dimension measurement between two points only requires calculation based on the relative distance between the two points, without needing to capture both points in the same field of view. Therefore, this invention uses multiple cameras to capture a portion of the module, performs camera calibration first, and then combines the circular hole information to obtain the transformation matrix of each camera in the same coordinate system, thus cleverly solving the problem encountered in the battery module dimension measurement mentioned in the background art.

[0076] In the multi-view visual ranging method for measuring the size of power battery modules provided in the embodiments of the present invention, a standard part can be placed at a fixed position on the logistics line to measure the size of the battery module, and calibration plates with different orientations can be placed on the surface of the standard part. Cameras with different orientations can be used to take corresponding pictures. Based on the captured images, the transformation matrices corresponding to different cameras can be mapped to the same coordinate system. Based on the transformation matrix in the same coordinate system and the corner point image coordinates of the battery module, the size of the battery module can be accurately obtained. Thus, it can be determined whether the size of the battery module is qualified within the allowable range of specifications, and the cost of use is low.

[0077] Furthermore, in a specific implementation, in the multi-view visual ranging method for measuring the size of the power battery module provided in the above embodiments of the present invention, step S103, obtaining the transformation matrix corresponding to each camera, may specifically include: first, extracting the image coordinates of the center of the circle on the captured calibration board image to form a center matrix; then, performing a transmission transformation between the physical matrix formed by the physical coordinates of the center of the circle on each calibration board and the center matrix to obtain the transformation matrix corresponding to each camera.

[0078] Specifically, each camera in each orientation captures an image of the calibration board. A 3x3 matrix (horizontal and vertical) containing the center of each circle is extracted to form a center matrix. This matrix, along with the physical coordinates of the calibration board containing the center of each circle, forms a physical matrix. A transmission transformation is then performed on these two matrices to obtain the transformation matrix corresponding to that camera. The calibration board images captured from the four orientations are shown below. Figures 5 to 8 As shown.

[0079] In specific implementation, the above steps extract the image coordinates of the center of the circle in the captured calibration board image to form a center matrix. Specifically, this may include: selecting a ROI rectangular area containing 3×3 circles for the captured calibration board image; calculating the center image coordinates in the ROI rectangular area; and sorting the calculated 3×3 center image coordinates in a set order to form a center matrix.

[0080] Figure 9 This shows a rectangular ROI region containing 3×3 circles, within which the image coordinates of the circle centers are calculated. Since many circles extracted based on blobs may be arranged in a "Z" or "N" pattern, such as... Figure 10 and Figure 11 As shown. In practical applications, other novel arrangements may also be used. To facilitate a one-to-one correspondence with the physical coordinates of the circle centers on the calibration board, the calculated coordinates of the 3×3 circle centers need to be sorted in a certain order.

[0081] In practice, the above steps involve sorting the image coordinates of the calculated 3×3 circle centers in a predetermined order, which may include:

[0082] First, based on the center point (Xc, Yc) and rotation angle θ of the ROI rectangular region, the ROI rectangular region is rotated, and the coordinates of the 3×3 circle centers are also rotated; the rotation formula is:

[0083]

[0084] Where (X,Y) are the coordinates before rotation, and (X',Y') are the coordinates after rotation.

[0085] Let (Xtl, Ytl), (Xtr, Ytr), (Xbr, Ybr), and (Xbl, Ybl) be the coordinates of the four vertices of the ROI rectangle, and (X'tl, Y'tl), (X'tr, Y'tr), (X'br, Y'br), and (X'bl, Y'bl) be the coordinates of the four vertices of the rotated rectangle. Figure 12 and 13 As shown.

[0086] Let (X1,Y1)~(X9,Y9) be the coordinates of the center of the 3×3 circle before rotation, and (X'1,Y'1)~(X'9,Y'9) be the coordinates of the center of the 3×3 circle after rotation.

[0087] Then, based on the coordinates of the vertices of the rotated rectangle, the rotated ROI rectangle is divided into 3×3 smaller rectangles. These 3×3 smaller rectangles are traversed sequentially in a "Z" pattern. When a center point (X'n, Y'n) from (X'1, Y'1) to (X'9, Y'9) is found to fall within a specific smaller rectangle, the corresponding (Xn, Yn) is placed in that position. This yields the coordinates of the 3×3 center points ordered in a "Z" pattern, denoted as (X1s, Y1s) to (X9s, Y9s). Figure 14 and Figure 15 As shown.

[0088] Where (X1s,Y1s)=(X1,Y1), (X2s,Y2s)=(X6,Y6), (X3s,Y3s)=(X7,Y7),

[0089] (X4s, Y4s) = (X2, Y2), (X5s, Y5s) = (X5, Y5), (X6s, Y6s) = (X8, Y8),

[0090] (X7s,Y7s)=(X3,Y3), (X8s,Y8s)=(X4,Y4), (X9s,Y9s)=(X9,Y9).

[0091] In addition, in specific implementation, the above steps involve performing a transmission transformation on the physical matrix composed of the physical coordinates of the centers of the circles on each calibration plate and the center matrix to obtain the transformation matrix corresponding to each camera. Specifically, this may include: calculating the offset angle between each calibration plate and the horizontal direction based on the 3×3 center image coordinates; obtaining the 3×3 corrected center physical coordinates of each calibration plate after rotating its physical coordinate system to be parallel to the image coordinate system based on the offset angle; and performing a transmission transformation on the 3×3 center image coordinates and the corresponding corrected center physical coordinates to obtain the transformation matrix corresponding to each camera.

[0092] Specifically, since the calibration plate may be offset from the horizontal direction, the offset angle can be calculated based on the coordinates of the 3×3 circle centers. The specific method is as follows:

[0093]

[0094]

[0095] Where α is the offset angle of the calibration plate relative to the horizontal direction, such as Figure 16 As shown.

[0096] The physical coordinate system containing the centers of the 3×3 circles is rotated to be parallel to the image coordinate system, i.e., the physical coordinates of the circle centers are corrected. The purpose is to ensure that the physical coordinate systems of the calibration plates captured by the four cameras (left front, left rear, right front, and right rear) are parallel, facilitating subsequent unified mapping to a single physical coordinate system. In practice, the following formula can be used to obtain the corrected physical coordinates of the 3×3 circle centers after the physical coordinate system of each calibration plate is rotated to be parallel to the image coordinate system:

[0097]

[0098] Where α represents the offset angle; (Xp, Yp) represents the physical coordinates of the 3×3 circle centers, which can be obtained from the center distance parameter of the calibration plate; (Xep, Yep) represents the physical coordinates of the 3×3 correction circle centers.

[0099] Table 1 shows the physical coordinate parameters of the calibration plate used in the experiment with a 3×3 "Z" shaped arrangement of the centers, and the corrected physical coordinates of the 3×3 centers.

[0100] Table 1 shows that the center distance between the horizontal and vertical circles on the calibration plate is 2mm.

[0101] Circle center index Pre-set physical coordinates (Xp, Yp) Corrected physical coordinates (Xep, Yep) 1 (0,0) (0,0) 2 (2,0) (1.99855,0.076021) 3 (4,0) (3.99711,0.152043) 4 (0,2) (-0.0760217,1.99855) 5 (2,2) (1.92253,2.07458) 6 (4,2) (3.92109,2.1506) 7 (0,4) (-0.152043,3.99711) 8 (2,4) (1.84651,4.07313) 9 (4,4) (3.84507,4.14915)

[0102] Next, the physical coordinates of the 3×3 corrected center points obtained from the "Z"-shaped arrangement are compared with the image coordinates of the 3×3 center points obtained from the "Z"-shaped arrangement to obtain the transformation matrix through perspective transformation. In practical implementation, the transformation matrix corresponding to each camera can be obtained using the following formula:

[0103]

[0104] The transformation matrix is ​​given by the following formula:

[0105]

[0106] Where A, B, D, and E represent rigid body transformations such as rotation, C and F represent horizontal and vertical translations respectively, and (Xs, Ys) represent the coordinates of the 3×3 circle centers. In fact, the six parameters A, B, C, D, E, and F can be solved using only six points. Here, the optimal solution for the six parameters can be obtained using the minimum variance method with 3×3 points.

[0107] The transformation matrices of the four cameras were obtained separately following the steps described above. Since these transformation matrices were obtained in different coordinate systems, even after correction, the coordinate systems are only parallel; they are not actually in the same coordinate system, which is detrimental to subsequent dimension calculations. For example, Table 1 sets the first center of the 3×3 circles on the calibration plate as the origin (Po-x, Po-y) of the physical coordinate system to unify the transformation matrices of the four cameras into the same coordinate system.

[0108] Therefore, in specific implementation, in the multi-view visual ranging method for measuring the size of the power battery module provided in the above embodiments of the present invention, step S104 corrects the transformation matrix to obtain the target transformation matrix in the same coordinate system. Specifically, it may include: extracting the center coordinates of the corresponding circular hole images for the center of the circular hole images in different orientations on the standard part; then using the transformation matrix and the center coordinates of the circular hole images to calculate the physical coordinates of the center of the circular hole in different orientations; then performing translation calculations on the preset starting point physical coordinates and the physical coordinates of the center of the circular hole in different orientations to correct the transformation matrix and obtain the target transformation matrix corresponding to the camera in different orientations in the same coordinate system.

[0109] In practical implementation, the above steps involve translating the preset starting point physical coordinates with the physical coordinates of the center of the circular holes in different orientations to correct the transformation matrix, thereby obtaining the target transformation matrix corresponding to the cameras in different orientations under the same coordinate system. Specifically, this may include: taking the physical coordinates of the center of the circular hole in one orientation as the origin of the physical coordinate system, calibrating the horizontal and vertical translation variables of the transformation matrix corresponding to the camera in that orientation, and obtaining the target transformation matrix corresponding to the camera in that orientation; then, based on the physical coordinates of the center of the circular holes in other orientations, calculating the distance from the center of the circular holes in other orientations to the origin of the physical coordinate system; and finally, based on the calculated distance from the center of the circular holes in other orientations to the origin of the physical coordinate system, obtaining the target transformation matrix corresponding to the cameras in other orientations.

[0110] Specifically, first, keeping the standard part stationary, remove the calibration plate on the front left side, and then take an image of the circular hole on the standard part, such as... Figure 17 As shown. Then, the coordinates of the center of the circular hole (Xph-tl, Yph-tl) are solved.

[0111] The physical coordinates corresponding to the center of the circular hole can be obtained using equation (6). The calculation formula is as follows:

[0112]

[0113] Where (Xwh-tl, Ywh-tl) represents the physical coordinates of the center of the circular hole on the left front standard part.

[0114] Then, by taking the point (Xwh-tl, Ywh-tl) as the origin (Po-x, Po-y) of the physical coordinate system, the translation variables of the transformation matrix corresponding to the left front camera can be calibrated:

[0115]

[0116] Where Csrc and Fsrc represent the lateral and longitudinal translation variables before calibration, and Cdst and Fdst represent the lateral and longitudinal translation variables after calibration, the calibration transformation matrix (i.e., the target transformation matrix) corresponding to the left front camera can be obtained according to equation (8):

[0117]

[0118] Similarly, the calibration plates were removed and images of the circular holes on the standard parts in the left rear, right front, and right rear positions were taken. The center coordinates (Xph-bl, Yph-bl), (Xph-tr, Yph-tr), and (Xph-br, Yph-br) of the corresponding circular hole images were extracted. Then, the physical coordinates of the center of the circular hole (Xwh-bl, Ywh-bl), (Xwh-tr, Ywh-tr), and (Xwh-br, Ywh-br) were calculated according to their respective transformation matrices. Since the parameters of the module standard parts are known, based on the known parameters such as the distances from the centers of the circular holes at the left rear, right front, and right rear positions to the center of the circular hole at the left front position, which are (ΔXbl-tl, ΔYbl-tl), (ΔXtr-tl, ΔYtr-tl), and (ΔXbr-tl, ΔYbr-tl), respectively, we can obtain the calibrated transformation matrices (i.e., target transformation matrices) for the left rear, right front, and right rear cameras when (Xwh-tl, Ywh-tl) is taken as the origin of the physical coordinate system:

[0119]

[0120] Equations (9), (10), (11), and (12) above are the target transformation matrices corresponding to the left front, left rear, right front, and right rear cameras, respectively.

[0121] Furthermore, in a specific implementation, in the multi-view visual ranging method for measuring the size of the power battery module provided in the embodiments of the present invention, step S105 extracts the image coordinates of the corner points of the battery module and combines them with the target transformation matrix to obtain the size of the battery module. Specifically, this may include: firstly, extracting the edges of the captured battery module image; then calculating the corner point image coordinates in different directions based on the extracted edges; then combining the corner point image coordinates in each direction with the corresponding target transformation matrix to calculate the physical coordinates of the corner points in each direction; finally, obtaining the size of the battery module based on the physical coordinates of the corner points in each direction.

[0122] Specifically, based on images of the battery module captured by four cameras from four different locations, the module's edges are extracted. The image coordinates of the corner points are calculated from these edges, resulting in four corner points: front left, rear left, front right, and rear right. Figure 18 As shown.

[0123] By combining the corner coordinates of each orientation with the corresponding calibration transformation matrix, and using an operation similar to Equation (7) to calculate the relative physical coordinates of the upper left circular hole, the relative physical coordinates can be obtained, which are respectively set as (Xo-tl,Yo-tl), (Xo-bl,Yo-bl), (Xo-tr,Yo-tr), and (Xo-br,Yo-br). Then, the length L of the module is defined by Equation (13), and the width W is defined by Equation (14):

[0124] L=((Xo-tr-Xo-tl)+(Xo-br-Xo-bl)) / 2 (13)

[0125] W=((Yo-bl-Yo-tl)+(Yo-br-Yo-tr)) / 2 (14)

[0126] The length and width measurements of the module are compared with the standard values. If they are within the allowable range of the specifications, the module is judged as a good product; otherwise, it is judged as a defective product.

[0127] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatuses, devices, and storage media disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the descriptions are relatively simple; relevant parts can be referred to the method section.

[0128] Those skilled in the art will further recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, computer software, or a combination of both. To clearly illustrate the interchangeability of hardware and software, the components and steps of the various examples have been generally described in terms of functionality in the foregoing description. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution. Those skilled in the art can use different methods to implement the described functions for each specific application, but such implementation should not be considered beyond the scope of this application.

[0129] The steps of the methods or algorithms described in conjunction with the embodiments disclosed herein can be implemented directly by hardware, a software module executed by a processor, or a combination of both. The software module can be located in random access memory (RAM), main memory, read-only memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art.

[0130] Finally, it should be noted that in this document, relational terms such as "first" and "second" are used only to distinguish one entity or operation from another, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Furthermore, the terms "comprising," "including," or any other variations thereof are intended to cover non-exclusive inclusion, such that a process, method, article, or apparatus that comprises a list of elements includes not only those elements but also other elements not expressly listed, or elements inherent to such a process, method, article, or apparatus. Without further limitations, an element defined by the phrase "comprising one..." does not exclude the presence of other identical elements in the process, method, article, or apparatus that includes said element.

[0131] The multi-view visual ranging method for measuring the size of power battery modules provided by this invention has been described in detail above. Specific examples have been used to illustrate the principle and implementation of this invention. The description of the above embodiments is only for the purpose of helping to understand the method and core idea of ​​this invention. At the same time, for those skilled in the art, there will be changes in the specific implementation and application scope based on the idea of ​​this invention. Therefore, the content of this specification should not be construed as a limitation of this invention.

Claims

1. A multi-view visual ranging method for measuring the size of a power battery module, characterized in that, include: A standard part is placed at a fixed position on the logistics line to measure the size of the battery module, and a calibration plate is placed on the surface of the standard part in the direction corresponding to the circular hole in the battery module. A camera corresponding to the calibration plate is placed at the same set height from the logistics line and in the position corresponding to the circular hole in the battery module. The calibration board is photographed using cameras in each direction, and the image coordinates of the center of the circle in the photographed calibration board image are extracted to form a center matrix; The offset angle between each calibration plate and the horizontal direction is calculated based on the coordinates of the 3×3 circle center images; Based on the offset angle, the physical coordinates of the centers of the 3×3 correction circles after the physical coordinate system of each calibration plate is rotated to be parallel to the image coordinate system are obtained; Perform a transmission transformation on the 3×3 center image coordinates and the corresponding corrected center physical coordinates to obtain the transformation matrix for each camera. Remove the calibration plate, take pictures of the standard part using cameras in each direction, and extract the center coordinates of the corresponding circular holes from the center of the circular hole images in different directions on the standard part. Using the transformation matrix and the center coordinates of the circular hole image, calculate the physical coordinates of the center of the circular hole in different orientations; The physical coordinates of the preset starting point are translated and calculated with the physical coordinates of the center of the circular holes in different directions to correct the transformation matrix and obtain the target transformation matrix corresponding to the camera in different directions in the same coordinate system. During online production on the logistics line, cameras from each direction are used to photograph the battery module, and the image coordinates of the corner points of the battery module are extracted and combined with the target transformation matrix to obtain the size of the battery module.

2. The multi-view visual ranging method for measuring the size of a power battery module according to claim 1, characterized in that, Extract the image coordinates of the center of the circle on the captured calibration board image to form a center matrix, including: For the captured calibration board image, select a rectangular ROI region containing 3×3 circles; Calculate the center image coordinates within the ROI rectangular region; The calculated coordinates of the 3×3 center images are sorted in a set order to form a center matrix.

3. The multi-view visual ranging method for measuring the size of a power battery module according to claim 2, characterized in that, Sort the calculated image coordinates of the 3×3 circle centers in a predetermined order, including: Based on the center point and rotation angle of the ROI rectangular region, the ROI rectangular region is rotated, and the coordinates of the 3×3 circle centers are also rotated. The rotated ROI rectangular region is divided into 3×3 small rectangles based on the coordinates of the vertices of the rotated rectangle. The 3×3 small rectangles are traversed sequentially in a Z-shape. When a circle center is found to fall within the nth small rectangle, the coordinates of that circle center are arranged in the nth position in the circle center matrix in a Z-shape, so as to obtain the coordinates of the 3×3 circle centers arranged in a Z-shape; n is a positive integer.

4. The multi-view visual ranging method for measuring the size of a power battery module according to claim 3, characterized in that, The following formula is used to obtain the physical coordinates of the 3×3 correction circle centers after the physical coordinate system of each calibration plate is rotated to be parallel to the image coordinate system: in, α Indicates the offset angle; XP , Yp () represents the physical coordinates of the centers of the 3×3 circles; Xep , Yep ) represents the physical coordinates of the 3×3 corrected center circles.

5. The multi-view visual ranging method for measuring the size of a power battery module according to claim 4, characterized in that, The transformation matrix for each camera is obtained using the following formula: in, Let A, B, D, and E represent rigid body transformations such as rotation, and C and F represent lateral and longitudinal translations, respectively. Xs , Ys ) represents the coordinates of the center of a 3×3 circle image.

6. The multi-view visual ranging method for measuring the size of a power battery module according to claim 5, characterized in that, The transformation matrix is ​​corrected by translating the preset starting point physical coordinates with the center physical coordinates of circular holes in different orientations to obtain the target transformation matrix corresponding to cameras in different orientations in the same coordinate system, including: Set the physical coordinates of the center of the circular hole in one of the directions as the origin of the physical coordinate system, calibrate the horizontal and vertical translation variables of the transformation matrix corresponding to the camera in that direction, and obtain the target transformation matrix corresponding to the camera in that direction. Based on the physical coordinates of the center of the circular holes in other directions, calculate the distance from the center of the circular holes in other directions to the origin of the physical coordinate system. Based on the calculated distances from the center of the circular aperture in other orientations to the origin of the physical coordinate system, the target transformation matrix corresponding to the camera in other orientations is obtained.

7. The multi-view visual ranging method for measuring the size of a power battery module according to claim 6, characterized in that, Extracting the image coordinates of the corner points of the battery module and combining them with the target transformation matrix yields the dimensions of the battery module, including: Edge extraction is performed on the captured images of the battery module; Calculate the corner coordinates of different orientations based on the extracted edges; Combine the corner image coordinates of each direction with the corresponding target transformation matrix to calculate the physical coordinates of the corner. The dimensions of the battery module are obtained based on the physical coordinates of the corner points in each direction.

Images