Visual-based keel bending and hollow recognition method and system

Abstract

The application relates to the field of building visual technology, and provides a keel bending and hollow identification method and system based on vision, which comprises the following steps: conveying the keel to an image acquisition position, acquiring the images of the top and end face of the keel through an image acquisition camera; identifying the keel bending and the keel hollow of the acquired images, and determining the sorting position of the keel according to the identification result; conveying the keel to a sorting conveying position, and conveying the keel to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined sorting position of the keel after the keel is in place. According to the scheme of the application, the bending steel wood keel can be visually identified by using the logical operation of the pixel value after image binarization, and the hollow steel wood keel can be visually identified by using the gray level co-occurrence matrix energy vector method. The hollow and bending steel wood keel can be automatically, efficiently and accurately identified through the image processing technology, and the sorting efficiency of the steel wood keel is improved.

Description

Technical Field

[0001] This invention relates to the field of architectural vision technology, and in particular to a vision-based method and system for identifying the bending and hollowing of keels. Background Technology

[0002] In the construction of cast-in-place concrete structures, keels play a crucial role in supporting and maintaining the shape of the concrete, enhancing structural strength and stability, and are used in enormous quantities on-site. Sorting and organizing are important steps in the recycling of construction keels. During the sorting process, keels with excessive bending or deformation, once entering the automated sorting production line, will clog the straightening machine during the straightening stage, significantly reducing the efficiency of the sorting process. Hollow keels have no recycling value; if they are not processed promptly, it will lead to wasted storage space and quality problems during reuse. Existing automated sorting equipment cannot automatically identify hollow keels, and manual work for extended periods can cause visual fatigue, resulting in missorting and missed inspections. Facing the monthly sorting demand of millions of square meters of keels in construction warehouses, relying on traditional manual identification and sorting methods has significant drawbacks, including high labor intensity, low efficiency, high error rate, and high labor costs. Summary of the Invention

[0003] The purpose of this invention is to solve at least one technical problem in the background art and to provide a vision-based method and system for identifying keel bending and hollowing.

[0004] To achieve the above objectives, the present invention provides a vision-based method for identifying keel bending and hollowness, comprising:

[0005] The keel is transmitted to the image acquisition position, and images of the top and end faces of the keel are acquired by the image acquisition camera.

[0006] The acquired images are used to identify keel bending and keel hollowness, and the sorting position of the keel is determined based on the identification results;

[0007] The keel is transported to the sorting and conveying position. After the keel is in place, it is transported to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined keel sorting position.

[0008] According to one aspect of the present invention, the keel is transported by a first conveying device, and image acquisition cameras are arranged at the top center position and on both the front and rear sides of the first conveying device.

[0009] When the image acquisition camera at the center of the first conveyor device captures the image acquisition position of the keel when it moves to the center of the first conveyor device, it issues a command, and the first conveyor device stops transmitting the keel according to the command.

[0010] Images of the top of the keel are captured by an image acquisition camera located at the center of the top of the first conveyor device, and images of the two ends of the keel are captured by image acquisition cameras located on the front and rear sides of the first conveyor device.

[0011] According to one aspect of the invention, the keel bending detection includes:

[0012] The acquired images are converted to grayscale.

[0013] The image after grayscale processing is then subjected to Gaussian filtering, median filtering, and binarization processing in sequence.

[0014] The processed image is then used to extract the keel edges using the Canny edge detection algorithm;

[0015] Remove small connected components from the extracted keel edge objects, then perform erosion, dilation, and filling on the image after removing the small connected components, and count the number of pixels in the eroded, dilated, and filled regions, denoted as N0. The value inside the region is set to 1, and the value outside the region is set to 0. The image is denoted as the first image.

[0016] Rotate the short side of the first image to a vertical orientation to obtain the second image. Set all pixels in the row containing the short side of the second image to 1 and the rest to 0 to obtain the third image.

[0017] Perform a pixel-wise AND operation between the second and third images to obtain the fourth image;

[0018] The second and fourth images are XORed by pixels to obtain the fifth image;

[0019] Count the number of pixels with a value of 1 in the fifth image, and denote it as N1;

[0020] Calculate p = N1 / N0, and compare p with the threshold p. c Comparison, if p <p c If the keel is not bent, and p > p c The keel is bent.

[0021] According to one aspect of the present invention, the hollow keel identification includes:

[0022] Image processing and background segmentation techniques are used to extract the end face image of the keel, and then grayscale processing is performed on the extracted end face image to obtain the grayscale distribution map of the end face of the tested keel.

[0023] The solid keel end face is extracted, and then the grayscale of the extracted solid end face image is processed to obtain the grayscale distribution map of the solid keel end face;

[0024] Using the gray-level co-occurrence matrix algorithm, the co-occurrence matrix is ​​solved for the gray-level distribution map of the end face of the tested keel and the gray-level distribution map of the solid keel end face, respectively. Then, each co-occurrence matrix is ​​normalized and the energy of each co-occurrence matrix is ​​calculated. By comparing the magnitude of the co-occurrence matrix energy values, it is determined whether the tested keel is hollow.

[0025] According to one aspect of the present invention, extracting the acquired end-face image of the keel includes:

[0026] Obtain the image of the keel end face, and then process the image into grayscale.

[0027] The Canny edge detection algorithm is used to extract the edge features of the keel. Then, the image after the edge features are extracted is subjected to erosion and dilation processing. Finally, the processed edges are connected to form a closed region. The closed region is filled with white and multiplied with the original image to achieve the extraction of the keel end face image.

[0028] According to one aspect of the present invention, the calculation of the energy of each co-occurrence matrix and the determination of whether the tested keel is hollow by comparing the magnitudes of the co-occurrence matrix energy values ​​include:

[0029] Calculate the four co-occurrence matrices of the segmented keel end face image, with a distance of 1 and angles of 0 degrees, 45 degrees, 90 degrees and 135 degrees respectively;

[0030] After calculating the four co-occurrence matrices of the segmented keel end face image, the energy of each co-occurrence matrix is ​​obtained. The four energy values ​​constitute an energy matrix, and the modulus of the four energy matrices is used as an element to form an energy vector. The modulus of the energy vector is then taken.

[0031] By judging the magnitude of the energy vector modulus of multiple segmented keel end face images, if the value of the energy vector modulus of any segmented image is lower than 90% of the average value, it is determined that the steel-wood keel has a hollow structure.

[0032] According to one aspect of the invention, the keel is transferred to a second conveying device aligned with it via a first conveying device, and an image acquisition camera is positioned at the center of the upper part of the second conveying device;

[0033] When the image acquisition camera positioned above the center of the second conveyor captures the keel's movement to the sorting and conveying position at the center of the second conveyor, it issues a command, and the second conveyor stops transmitting the keel according to the command.

[0034] The second conveying device transports the keel to the good product temporary storage location, the curved temporary storage location, or the hollow temporary storage location according to the determined keel sorting location.

[0035] To achieve the above objectives, the present invention also provides a vision-based keel bending and hollowing identification system, comprising:

[0036] The image acquisition module transmits the keel to the image acquisition position, and the image acquisition camera captures images of the top and end faces of the keel.

[0037] The sorting location determination module performs keel bending and hollow identification on the acquired images, and determines the sorting location of the keel based on the identification results;

[0038] The keel conveying module transports the keel to the sorting and conveying position. After the keel is in place, it is transported to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined keel sorting position.

[0039] To achieve the above objectives, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the vision-based keel bending and hollow recognition method as described above.

[0040] To achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the vision-based keel bending and hollow recognition method described above.

[0041] This invention utilizes logical operations on pixel values ​​after image binarization to perform visual recognition of curved steel-wood keel, and employs the gray-level co-occurrence matrix energy vector method to perform visual recognition of hollow steel-wood keel. This invention, through image processing technology, enables automatic, efficient, and accurate identification of hollow and curved steel-wood keel, improving the sorting efficiency of steel-wood keel. Attached Figure Description

[0042] Figure 1 A flowchart illustrating a vision-based keel bending and hollowing identification method according to an embodiment of the present invention;

[0043] Figure 2 This is a front view of the device structure for keel transportation and identification detection in Example 1;

[0044] Figure 3 This is a top view of the device structure for keel transportation and identification detection in Example 1;

[0045] Figure 4 A diagram illustrating the principle of keel bending. Detailed Implementation

[0046] The invention will now be discussed with reference to exemplary embodiments. It should be understood that the described embodiments are merely intended to enable those skilled in the art to better understand and thus implement the invention, and are not intended to imply any limitation on the scope of the invention.

[0047] As used herein, the term "comprising" and its variations are to be interpreted as open-ended terms meaning "including but not limited to". The term "based on" is to be interpreted as "at least partially based on". The terms "one embodiment" and "an embodiment" are to be interpreted as "at least one embodiment".

[0048] Figure 1 The flowchart schematically illustrates a vision-based method for identifying keel bending and hollowness according to an embodiment of the present invention. Figure 1 As shown, in this embodiment, the vision-based keel bending and hollow recognition method includes:

[0049] The keel is transmitted to the image acquisition position, and images of the top and end faces of the keel are acquired by the image acquisition camera.

[0050] The acquired images are used to identify keel bending and keel hollowness, and the sorting position of the keel is determined based on the identification results;

[0051] The keel is transported to the sorting and conveying position. After the keel is in place, it is transported to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined keel sorting position.

[0052] Furthermore, according to one embodiment of the present invention, the keel is transported by a first conveying device, and image acquisition cameras are arranged at the top center position and on both the front and rear sides of the first conveying device.

[0053] When the image acquisition camera at the center of the first conveyor device captures the image acquisition position of the keel when it moves to the center of the first conveyor device, it issues a command, and the first conveyor device stops transmitting the keel according to the command.

[0054] Images of the top of the keel are captured by an image acquisition camera located at the center of the top of the first conveyor device, and images of the two ends of the keel are captured by image acquisition cameras located on the front and rear sides of the first conveyor device.

[0055] Furthermore, according to one embodiment of the present invention, keel bending identification includes:

[0056] The acquired images are converted to grayscale.

[0057] The image after grayscale processing is then subjected to Gaussian filtering, median filtering, and binarization processing in sequence.

[0058] The processed image is then used to extract the keel edges using the Canny edge detection algorithm;

[0059] Remove small connected components from the extracted keel edge objects, then perform erosion, dilation, and filling on the image after removing the small connected components, and count the number of pixels in the eroded, dilated, and filled region (i.e. the region surrounded by the edge detection), denoted as N0. The region inside the region is set to 1, and the region outside the region is set to 0. The image is denoted as the first image.

[0060] Rotate the short side of the first image to a vertical orientation to obtain the second image. Set all pixels in the row containing the short side of the second image to 1 and the rest to 0 to obtain the third image.

[0061] Perform a pixel-wise AND operation between the second and third images to obtain the fourth image;

[0062] The second and fourth images are XORed by pixels to obtain the fifth image;

[0063] Count the number of pixels with a value of 1 in the fifth image, and denote it as N1;

[0064] Calculate p = N1 / N0, compare p with 10%. If p < 10%, the keel is not bent; if p > 10%, the keel is bent.

[0065] Furthermore, according to one embodiment of the present invention, the keel hollow identification includes:

[0066] Image processing and background segmentation techniques are used to extract the end face image of the keel, and then grayscale processing is performed on the extracted end face image to obtain the grayscale distribution map of the end face of the tested keel.

[0067] The solid keel end face is extracted, and then the extracted end face image is processed in grayscale, compressing 256 grayscale levels to 16 levels. The grayscale image of the keel end face is evenly divided into 8 small images using image segmentation.

[0068] Using the gray-level co-occurrence matrix algorithm, the gray-level co-occurrence matrices in four directions (0°, 45°, 90°, and 135°) are calculated for each of the eight segmentation images. Then, each co-occurrence matrix is ​​normalized, and the energy of the four co-occurrence matrices for each image is calculated. The energy of the four energy matrices is used as elements to form an energy vector. The modulus of the energy vector is taken, and the value of the modulus of the energy vector of the eight segmentation images is used to determine whether there is a hollow steel-wood keel. If the value of the modulus of the energy vector of any segmentation image is lower than 90% of the average value, it is determined that there is a hollow steel-wood keel.

[0069] Furthermore, according to one embodiment of the present invention, the keel is transferred to a second conveying device aligned with it via a first conveying device, and an image acquisition camera is positioned at the center of the upper part of the second conveying device.

[0070] When the image acquisition camera positioned above the center of the second conveyor captures the keel's movement to the sorting and conveying position at the center of the second conveyor, it issues a command, and the second conveyor stops transmitting the keel according to the command.

[0071] The second conveying device transports the keel to the good product temporary storage location, the curved temporary storage location, or the hollow temporary storage location according to the determined keel sorting location.

[0072] Furthermore, to achieve the above objectives, the present invention also provides a vision-based keel bending and hollowing identification system, comprising:

[0073] The image acquisition module transmits the keel to the image acquisition position, and the image acquisition camera captures images of the top and end faces of the keel.

[0074] The sorting location determination module performs keel bending and hollow identification on the acquired images, and determines the sorting location of the keel based on the identification results;

[0075] The keel conveying module transports the keel to the sorting and conveying position. After the keel is in place, it is transported to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined keel sorting position.

[0076] The vision-based keel bending and hollow recognition system of the present invention can realize the vision-based keel bending and hollow recognition method. The specific process steps are as described above and will not be repeated here.

[0077] Furthermore, to achieve the above objectives, the present invention also provides an electronic device, including a processor, a memory, and a computer program stored in the memory and executable on the processor. When the computer program is executed by the processor, it implements the vision-based keel bending and hollow recognition method described above.

[0078] Furthermore, to achieve the above objectives, the present invention also provides a computer-readable storage medium storing a computer program, which, when executed by a processor, implements the vision-based keel bending and hollow recognition method described above.

[0079] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely one preferred embodiment of the invention and are only used to explain the invention. They do not limit the scope of protection of the invention. All other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this invention.

[0080] Example 1

[0081] A vision-based method for identifying keel curvature and hollowness includes the following steps:

[0082] S1: As Figure 2 and Figure 3 As shown, the first conveyor device 1 transports the keel 2. When the industrial camera 4 detects that the keel has reached the middle of the movable first conveyor device 1, it issues a command, and the first conveyor device 1 stops transporting the keel. The industrial cameras 3, 4, and 5 acquire images of the top and end faces of the keel 2, and then perform keel bending and hollow recognition on the images to determine the final stopping position (sorting position) of the second conveyor device 7. After the image recognition is completed, it is necessary to ensure that the movable second conveyor device 7 is aligned with the first conveyor device 1, and then start the first conveyor device 1 to transport the keel 2 onto the second conveyor device 7.

[0083] S2: Monitor the keel on the movable second conveyor 7. When the industrial camera 6 detects that the keel has moved to the middle of the movable second conveyor 7, it issues a command to stop the movable second conveyor 7 from conveying. Then, based on the information of the docking position obtained in step S1, the second conveyor 7 moves to the target position (i.e., the docking position).

[0084] S3: Activate the movable second conveyor 7 to convey the keel to the third conveyor 8 (good product temporary storage position), the fourth conveyor 9 (bending temporary storage position), or the fifth conveyor 10 (hollow temporary storage position). After an appropriate time (when the keel has completely moved to the third conveyor 8, the fourth conveyor 9, or the fifth conveyor 10), return the movable second conveyor 7 to its original position (aligned with the first conveyor 1).

[0085] S4: Repeat S1~S3 in sequence to sort the keel.

[0086] like Figure 4 As shown, in this embodiment, keel bending identification includes:

[0087] S1: Perform image grayscale processing on the acquired image;

[0088] S2: Perform Gaussian filtering, median filtering and binarization on the grayscale image obtained in S1 in sequence;

[0089] S3: Extract the keel edge from the image processed in S2 using the Canny algorithm;

[0090] S4: Remove small objects from the edge objects obtained from S3, perform erosion, dilation and filling, and count the number of pixels in the region, denoted as N0. The value is set to 1 within the region and 0 outside the region. The image is denoted as imag_0.

[0091] S5: Rotate the short side of the image obtained in S4 to be vertical, and denote the image as imag_1. Set all pixels in the row containing the short side pixels of imag_1 to 1, and set the rest to 0, imag_2;

[0092] S6: Perform a pixel-wise AND operation on imag_1 and imag_2 to obtain imag_3;

[0093] S7: Perform an XOR operation on imag_1 and imag_3 by pixels to obtain imag_4;

[0094] S8: Count the number of pixels with a value of 1 in imag_4, and denote it as N1;

[0095] S9: Calculate p=N1 / N0, compare p with 10%. If p<10%, the keel is not bent; if p>10%, the keel is bent.

[0096] In this embodiment, the hollow keel identification includes:

[0097] T1: Using image processing and background segmentation techniques, the acquired keel end face image is extracted and grayscale processed to obtain the grayscale distribution map of the tested keel end face;

[0098] T2: Extract the solid keel end face, then perform grayscale processing on the extracted end face image, compressing 256 grayscale levels to 16 levels, and using image segmentation to evenly divide the grayscale image of the keel end face into 8 small images.

[0099] T3: Using the gray-level co-occurrence matrix algorithm, the gray-level co-occurrence matrices in four directions (0°, 45°, 90°, and 135°) are solved for the eight segmentation images respectively. Then, each co-occurrence matrix is ​​normalized, and the energy of the four co-occurrence matrices for each image is calculated. The energy of the four energy matrices is used as elements to form an energy vector. The modulus of the energy vector is taken, and the value of the modulus of the energy vector of the eight segmentation images is used to determine whether there is a segmentation image whose modulus of the energy vector is lower than 90% of the average value. If so, it is determined that the steel-wood keel has a hollow structure.

[0100] In step T1, keel end face extraction includes acquiring the keel end face image, then performing grayscale processing on the image, using the Canny edge detection algorithm to extract keel edge features, then performing erosion and dilation processing on the edge-detected image, and finally connecting the processed edges to form a closed region, filling the closed region with white and performing a dot product with the original image to achieve keel end face extraction.

[0101] In step T3, four co-occurrence matrices are calculated for the segmented image, with a distance of 1 and angles of 0, 45, 90, and 135 degrees. The magnitudes of the energy vectors of the eight segmented images are evaluated; if any segmented image has an energy vector magnitude lower than 90% of the average, then the steel-wood keel is determined to have hollow sections.

[0102] Those skilled in the art will recognize that the modules and algorithm steps described in conjunction with the embodiments disclosed herein can be implemented using electronic hardware, or a combination of computer software and electronic hardware. 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 implementations should not be considered beyond the scope of this invention.

[0103] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working process of the above-described apparatus and equipment can be referred to the corresponding process in the foregoing method implementation, and will not be repeated here.

[0104] In the embodiments provided in this application, it should be understood that the disclosed apparatus and methods can be implemented in other ways. For example, the apparatus embodiments described above are merely illustrative; for instance, the division of modules is only a logical functional division, and in actual implementation, there may be other division methods. For example, multiple modules or components may be combined or integrated into another system, or some features may be ignored or not executed. Furthermore, the coupling or direct coupling or communication connection shown or discussed may be through some interfaces; the indirect coupling or communication connection between apparatuses or modules may be electrical, mechanical, or other forms.

[0105] The modules described as separate components may or may not be physically separate. The components shown as modules may or may not be physical modules; that is, they may be located in one place or distributed across multiple network modules. Some or all of the modules can be selected to achieve the objectives of the embodiments of the present invention, depending on actual needs.

[0106] In addition, the functional modules in the embodiments of the present invention can be integrated into one processing module, or each module can exist physically separately, or two or more modules can be integrated into one module.

[0107] If the aforementioned functions are implemented as software functional modules and sold or used as independent products, they can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, essentially, or the part that contributes to the prior art, or a portion of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the sending / receiving methods of various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, ROM, RAM, magnetic disks, or optical disks.

[0108] The above description is merely a preferred embodiment of this application and an explanation of the technical principles employed. Those skilled in the art should understand that the scope of the invention involved in this application is not limited to technical solutions formed by specific combinations of the above-described technical features, but should also cover other technical solutions formed by arbitrary combinations of the above-described technical features or their equivalents without departing from the inventive concept. For example, technical solutions formed by substituting the above-described features with (but not limited to) technical features with similar functions disclosed in this application.

[0109] It should be understood that the sequence number of each step in the invention and its embodiments does not absolutely imply the order of execution. The execution order of each process should be determined by its function and internal logic, and should not constitute any limitation on the implementation process of the embodiments of the present invention.

Claims

1. A vision-based method for identifying keel bending and hollowness, characterized in that, include: The keel is transmitted to the image acquisition position, and images of the top and end faces of the keel are acquired by the image acquisition camera. The acquired images are used to identify keel bending and keel hollowness, and the sorting position of the keel is determined based on the identification results; The keel is transported to the sorting and conveying position. After the keel is in place, it is transported to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined keel sorting position. The keel is transported through the first conveying device, and image acquisition cameras are installed at the center above the first conveying device and on both the front and rear sides. When the image acquisition camera at the center of the first conveyor device captures the image acquisition position of the keel when it moves to the center of the first conveyor device, it issues a command, and the first conveyor device stops transmitting the keel according to the command. The image acquisition camera at the center of the top of the keel is used to acquire images of the top of the keel, and the image acquisition cameras on the front and rear sides of the first conveyor are used to acquire images of the two end faces of the keel. The hollow keel identification includes: Image processing and background segmentation techniques are used to extract the end face image of the keel, and then grayscale processing is performed on the extracted end face image to obtain the grayscale distribution map of the end face of the tested keel. The solid keel end face is extracted, and then the grayscale of the extracted solid end face image is processed to obtain the grayscale distribution map of the solid keel end face; Using the gray-level co-occurrence matrix algorithm, the co-occurrence matrix is ​​solved for the gray-level distribution map of the end face of the tested keel and the gray-level distribution map of the solid keel end face, respectively. Then, each co-occurrence matrix is ​​normalized and the energy of each co-occurrence matrix is ​​calculated. By comparing the magnitude of the co-occurrence matrix energy values, it is determined whether the tested keel has a hollow structure. The keel is transferred from the first conveyor to the second conveyor aligned with it, and an image acquisition camera is placed at the center of the upper part of the second conveyor. When the image acquisition camera positioned above the center of the second conveyor captures the keel's movement to the sorting and conveying position at the center of the second conveyor, it issues a command, and the second conveyor stops transmitting the keel according to the command. The second conveying device transports the keel to the good product temporary storage location, the curved temporary storage location, or the hollow temporary storage location according to the determined keel sorting location.

2. The vision-based keel bending and hollow recognition method according to claim 1, characterized in that, The keel bending recognition includes: The acquired images are converted to grayscale. The image after grayscale processing is then subjected to Gaussian filtering, median filtering, and binarization processing in sequence. The processed image is then used to extract the keel edges using the Canny edge detection algorithm; Remove small connected components from the extracted keel edge objects, then perform erosion, dilation, and filling on the image after removing the small connected components, and count the number of pixels in the eroded, dilated, and filled regions, denoted as N0. The value inside the region is set to 1, and the value outside the region is set to 0. The image is denoted as the first image. Rotate the short side of the first image to a vertical orientation to obtain the second image. Set all pixels in the row containing the short side of the second image to 1 and the rest to 0 to obtain the third image. Perform a pixel-wise AND operation between the second and third images to obtain the fourth image; The second and fourth images are XORed by pixels to obtain the fifth image; Count the number of pixels with a value of 1 in the fifth image, and denote it as N1; Calculate p = N1 / N0, and compare p with the threshold p. c Comparison, if p <p c If the keel is not bent, and p > p c The keel is curved.

3. The vision-based keel bending and hollow recognition method according to claim 1, characterized in that, The acquired end-face images of the keel are extracted, including: Obtain the image of the keel end face, and then process the image into grayscale. The Canny edge detection algorithm is used to extract the edge features of the keel. Then, the image after the edge features are extracted is subjected to erosion and dilation processing. Finally, the processed edges are connected to form a closed region. The closed region is filled with white and multiplied with the original image to achieve the extraction of the keel end face image.

4. The vision-based keel bending and hollow recognition method according to claim 1, characterized in that, The calculation of the energy of each co-occurrence matrix, by comparing the magnitude of the co-occurrence matrix energy values, determines whether the tested keel has hollow parts. Calculate the four co-occurrence matrices of the segmented keel end face image, with a distance of 1 and angles of 0 degrees, 45 degrees, 90 degrees and 135 degrees respectively; After calculating the four co-occurrence matrices of the segmented keel end face image, the energy of each co-occurrence matrix is ​​obtained. The four energy values ​​constitute an energy matrix, and the modulus of the four energy matrices is used as an element to form an energy vector. The modulus of the energy vector is then taken. By judging the magnitude of the energy vector modulus of multiple segmented keel end face images, if the value of the energy vector modulus of any segmented image is lower than 90% of the average value, it is determined that the steel-wood keel has a hollow structure.

5. A vision-based keel bending and hollowness recognition system, characterized in that, include: The image acquisition module transmits the keel to the image acquisition position, and the image acquisition camera captures images of the top and end faces of the keel. The sorting location determination module performs keel bending and hollow identification on the acquired images, and determines the sorting location of the keel based on the identification results; The keel conveying module transports the keel to the sorting and conveying position. After the keel is in place, it transports the keel to the corresponding good product temporary storage position, bending temporary storage position or hollow temporary storage position according to the determined keel sorting position. The keel is transported through the first conveying device, and image acquisition cameras are installed at the center above the first conveying device and on both the front and rear sides. When the image acquisition camera at the center of the first conveyor device captures the image acquisition position of the keel when it moves to the center of the first conveyor device, it issues a command, and the first conveyor device stops transmitting the keel according to the command. The image acquisition camera at the center of the top of the keel is used to acquire images of the top of the keel, and the image acquisition cameras on the front and rear sides of the first conveyor are used to acquire images of the two end faces of the keel. The hollow keel identification includes: Image processing and background segmentation techniques are used to extract the end face image of the keel, and then grayscale processing is performed on the extracted end face image to obtain the grayscale distribution map of the end face of the tested keel. The solid keel end face is extracted, and then the grayscale of the extracted solid end face image is processed to obtain the grayscale distribution map of the solid keel end face; Using the gray-level co-occurrence matrix algorithm, the co-occurrence matrix is ​​solved for the gray-level distribution map of the end face of the tested keel and the gray-level distribution map of the solid keel end face, respectively. Then, each co-occurrence matrix is ​​normalized and the energy of each co-occurrence matrix is ​​calculated. By comparing the magnitude of the co-occurrence matrix energy values, it is determined whether the tested keel has a hollow structure. The keel is transferred from the first conveyor to the second conveyor aligned with it, and an image acquisition camera is placed at the center of the upper part of the second conveyor. When the image acquisition camera positioned above the center of the second conveyor captures the keel's movement to the sorting and conveying position at the center of the second conveyor, it issues a command, and the second conveyor stops transmitting the keel according to the command. The second conveying device transports the keel to the good product temporary storage location, the curved temporary storage location, or the hollow temporary storage location according to the determined keel sorting location.

6. An electronic device, characterized in that, The method includes a processor, a memory, and a computer program stored in the memory and executable on the processor, wherein the computer program, when executed by the processor, implements the vision-based keel bending and hollow recognition method as described in any one of claims 1-4.

7. A computer-readable storage medium, characterized in that, A computer program is stored on the computer-readable storage medium, which, when executed by a processor, implements the vision-based keel bending and hollow recognition method as described in any one of claims 1-4.

Images