Steel coil centering identification method and device, electronic equipment and storage medium

Abstract

The application provides a steel coil centering recognition method and device, electronic equipment and storage medium. The method comprises the following steps: acquiring a current steel coil production image, the current steel coil production image comprising a steel coil and a coil carrying vehicle carrying the steel coil; performing target detection on the current steel coil production image to obtain a detection result, the detection result comprising current position information of the steel coil and current position information of the coil carrying vehicle; determining a separation state of the steel coil and the coil carrying vehicle based on the current position information of the steel coil and the current position information of the coil carrying vehicle; if the separation state of the steel coil and the coil carrying vehicle is separated, determining a centering state of the steel coil according to the current position information of the steel coil; and if the centering state of the steel coil is abnormal, performing alarm. The method can realize automatic steel coil centering recognition, timely alarm in the case of abnormal centering, avoid safety accidents, avoid human factor interference, improve the accuracy, stability and efficiency of steel coil centering recognition, and reduce the workload of relevant personnel.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of steel coil detection, and in particular to a steel coil centering recognition method and device, an electronic device and a storage medium. BACKGROUND

[0002] In the production of steel products, a steel coil is a common intermediate form of steel material. The steel coil is usually processed into a long strip shape through a continuous hot rolling process and then wound into a cylindrical roll for further processing. If an abnormal centering of the steel coil occurs in the production site of the steel coil conveying, a safety production accident such as a roll-over may occur, which may cause damage to equipment, or even personal injury or death.

[0003] At present, the steel coil centering recognition mainly relies on manual recognition by experienced workers. Since there are multiple steel coil production lines and each steel coil production line operates for a long time, the workload of manual recognition is large and the workers are prone to fatigue. In addition, the manual recognition is not stable enough and is prone to large recognition errors and recognition omissions, which may lead to the failure to timely discover the abnormal centering and cause safety accidents. SUMMARY

[0004] In view of the above-mentioned shortcomings of the prior art, the present application provides a steel coil centering recognition method, device, electronic device and storage medium to solve the technical problems of large workload, insufficient stability, large recognition errors and recognition omissions in manual steel coil centering recognition.

[0005] The present application provides a steel coil centering recognition method, which comprises: acquiring a current steel coil production image, the current steel coil production image comprising a steel coil and a coil conveying vehicle carrying the steel coil; performing target detection on the current steel coil production image to obtain a detection result, the detection result comprising current position information of the steel coil and current position information of the coil conveying vehicle; determining a separation state of the steel coil and the coil conveying vehicle based on the current position information of the steel coil and the current position information of the coil conveying vehicle, and if the separation state of the steel coil and the coil conveying vehicle is separated, determining a centering state of the steel coil according to the current position information of the steel coil; and if the centering state of the steel coil is abnormal, performing an alarm.

[0006] In an embodiment of the present application, before determining the separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier, the method comprises: comparing the current position information of the steel coil with preset region-of-interest position information of a preset region of interest to determine a relative position relationship between the steel coil and the preset region of interest; comparing the current position information of the coil carrier with the preset region-of-interest position information to determine a relative position relationship between the coil carrier and the preset region of interest; and if the steel coil and the coil carrier are both located in the preset region of interest, determining the separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier.

[0007] In an embodiment of the present application, determining the separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier comprises: comparing a current top-left longitudinal coordinate of a detection frame corresponding to the coil carrier with a current bottom-right longitudinal coordinate of a detection frame corresponding to the steel coil to obtain a first comparison result, wherein the current position information of the coil carrier comprises the current top-left longitudinal coordinate of the detection frame corresponding to the coil carrier, and the current position information of the steel coil comprises the current bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil; comparing a current bottom-right transverse coordinate of the detection frame corresponding to the coil carrier with a preset center line transverse coordinate to obtain a second comparison result, wherein the current position information of the coil carrier further comprises the current bottom-right transverse coordinate of the detection frame corresponding to the coil carrier; calculating a difference between an initial bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil and the current bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil as a first difference, wherein the initial bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil is obtained based on target detection on an initial steel coil production image; calculating a difference between the current top-left longitudinal coordinate of the detection frame corresponding to the coil carrier and the current bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil as a second difference; if the first comparison result satisfies a first preset condition, and the second comparison result satisfies a second preset condition, and the first difference is greater than a first preset difference, and the second difference is greater than a second preset difference, the separation state of the steel coil and the coil carrier is separated; if the first comparison result does not satisfy the first preset condition, and / or the second comparison result does not satisfy the second preset condition, and / or the first difference is less than or equal to the first preset difference, and / or the second difference is less than or equal to the second preset difference, the separation state of the steel coil and the coil carrier is not separated.

[0008] In an embodiment of the present application, the centering state of the steel coil is determined according to the current position information of the steel coil, which includes: calculating the current center horizontal coordinate of the detection frame corresponding to the steel coil according to the current left upper corner horizontal coordinate of the detection frame corresponding to the steel coil and the current right lower corner horizontal coordinate of the detection frame corresponding to the steel coil; calculating the difference between the current center horizontal coordinate of the detection frame corresponding to the steel coil and the preset center line horizontal coordinate as the initial offset of the steel coil, and determining the actual offset of the steel coil according to the initial offset of the steel coil and a preset proportion; if the actual offset of the steel coil is less than or equal to a preset offset, the centering state of the steel coil is normal; if the actual offset of the steel coil is greater than the preset offset, the centering state of the steel coil is abnormal.

[0009] In an embodiment of the present application, after determining the relative position relationship between the coil carrier and the preset region of interest, the method further includes: when there are multiple steel coils and the coil carrier carrying each steel coil is located in the preset region of interest, calculating the current area of the detection frame corresponding to each steel coil according to the current corner point coordinate information of the detection frame corresponding to each steel coil, the current position information of the steel coil including the current corner point coordinate information of the detection frame corresponding to the steel coil; if the current area of the detection frame corresponding to a steel coil meets the preset camera near point corresponding area interval, the steel coil is taken as a target steel coil, and the coil carrier carrying the steel coil is taken as a target coil carrier, and the separation state of the target steel coil and the target coil carrier is determined based on the current position information of the target steel coil and the current position information of the target coil carrier.

[0010] In an embodiment of the present application, if the current areas of all the detection frames corresponding to the steel coils do not meet the preset camera near point corresponding area interval, the method includes: comparing the current area of the detection frame corresponding to each steel coil with the preset camera far point corresponding area interval; if the current area of the detection frame corresponding to a steel coil meets the preset camera far point corresponding area interval, the steel coil is taken as a target steel coil, and the coil carrier carrying the steel coil is taken as a target coil carrier, and the separation state of the target steel coil and the target coil carrier is determined based on the current position information of the target steel coil and the current position information of the target coil carrier.

[0011] In an embodiment of the present application, determining the separation state of the target steel coil and the target coil handling vehicle based on the current position information of the target steel coil and the current position information of the target coil handling vehicle comprises: comparing the current top-left longitudinal coordinate of the detection frame corresponding to the target coil handling vehicle with the current bottom-right longitudinal coordinate of the detection frame corresponding to the target steel coil to obtain a first comparison result, wherein the current position information of the target coil handling vehicle comprises the current top-left longitudinal coordinate of the detection frame corresponding to the target coil handling vehicle, and the current position information of the target steel coil comprises the current bottom-right longitudinal coordinate of the detection frame corresponding to the target steel coil; comparing the current bottom-right lateral coordinate of the detection frame corresponding to the target coil handling vehicle with a preset center line lateral coordinate to obtain a second comparison result, wherein the current position information of the target coil handling vehicle further comprises the current bottom-right lateral coordinate of the detection frame corresponding to the target coil handling vehicle; calculating the difference between the initial bottom-right longitudinal coordinate of the detection frame corresponding to the target steel coil and the current bottom-right longitudinal coordinate of the detection frame corresponding to the target steel coil as a first difference value, wherein the initial bottom-right longitudinal coordinate of the detection frame corresponding to the target steel coil is obtained based on target detection on an initial steel coil production image; calculating the difference between the current top-left longitudinal coordinate of the detection frame corresponding to the target coil handling vehicle and the current bottom-right longitudinal coordinate of the detection frame corresponding to the target steel coil as a second difference value; if the first comparison result satisfies a first preset condition, and the second comparison result satisfies a second preset condition, and the first difference value is greater than a first preset difference value, and the second difference value is greater than a second preset difference value, then the separation state of the target steel coil and the target coil handling vehicle is separated; if the first comparison result does not satisfy the first preset condition, and / or the second comparison result does not satisfy the second preset condition, and / or the first difference value is less than or equal to the first preset difference value, and / or the second difference value is less than or equal to the second preset difference value, then the separation state of the target steel coil and the target coil handling vehicle is not separated.

[0012] In an embodiment of the present application, if the separation state of the target steel coil and the target coil carrier is separated, the method comprises: calculating a current center horizontal coordinate of a detection frame corresponding to the target steel coil according to a current left upper corner horizontal coordinate of the detection frame corresponding to the target steel coil and a current right lower corner horizontal coordinate of the detection frame corresponding to the target steel coil, wherein the current position information of the target steel coil comprises the current left upper corner horizontal coordinate of the detection frame corresponding to the target steel coil and the current right lower corner horizontal coordinate of the detection frame corresponding to the target steel coil; calculating a difference between the current center horizontal coordinate of the detection frame corresponding to the target steel coil and the preset center line horizontal coordinate as an initial offset of the target steel coil, and determining an actual offset of the target steel coil according to the initial offset of the target steel coil and a preset proportion; if the actual offset of the target steel coil is less than or equal to a preset offset, the centering state of the target steel coil is normal; if the actual offset of the target steel coil is greater than the preset offset, the centering state of the target steel coil is abnormal.

[0013] In an embodiment of the present application, before target detection is performed on the current steel coil production image, the method comprises: obtaining a plurality of historical steel coil production images, each historical steel coil production image comprising a historical steel coil and a historical coil carrier carrying the historical steel coil; performing data labeling on the historical steel coil and the historical coil carrier in each historical steel coil production image to obtain a sample data set; and training an initial target detection model using the sample data set to obtain a trained target detection model, so as to perform target detection on the current steel coil production image by using the trained target detection model.

[0014] In an embodiment of the present application, a steel coil centering recognition device is also provided, which comprises: an acquisition module configured to acquire a current steel coil production image, the current steel coil production image comprising a steel coil and a coil carrier carrying the steel coil; a detection module configured to perform target detection on the current steel coil production image to obtain a detection result, the detection result comprising current position information of the steel coil and current position information of the coil carrier; a determination module configured to determine a separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier, and determine a centering state of the steel coil according to the current position information of the steel coil if the separation state of the steel coil and the coil carrier is separated; and an alarm module configured to perform alarm if the centering state of the steel coil is abnormal.

[0015] In an embodiment of the present application, an electronic device is also provided, which comprises: one or more processors; and a storage device configured to store one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the steel coil centering recognition method as described above.

[0016] In an embodiment of the present application, a computer readable storage medium is also provided, which stores a computer program. When the computer program is executed by a processor of a computer, the computer is caused to execute the steel coil centering recognition method as described above.

[0017] The present application provides a steel coil centering recognition method, device, electronic equipment and storage medium. The method detects a target in a current steel coil production image to obtain current position information of the steel coil and current position information of a coil carrier. Whether the steel coil is separated from the coil carrier is determined based on the current position information of the steel coil and the current position information of the coil carrier. When the steel coil is separated from the coil carrier, whether the steel coil is abnormally centered is determined according to the current position information of the steel coil. If the steel coil is abnormally centered, an alarm is given. The method automatically recognizes the centering of the steel coil, avoids human interference, improves the accuracy and stability of the steel coil centering recognition, gives an alarm in time when the steel coil is abnormally centered, and thus avoids safety accidents. The method can also replace manual steel coil centering recognition, effectively improves the efficiency of the steel coil centering recognition, and reduces the workload of relevant personnel.

[0018] It should be understood that the foregoing general description and the following detailed description are only exemplary and explanatory, and are not limiting to the present application. BRIEF DESCRIPTION OF DRAWINGS

[0019] Figure 1 is a schematic diagram of an implementation environment of a steel coil centering recognition method according to an example embodiment of the present application;

[0020] Figure 2 is a flowchart of a steel coil centering recognition method according to an example embodiment of the present application;

[0021] Figure 3 is a schematic diagram of the positions of a steel coil and a coil carrier according to an example embodiment of the present application;

[0022] Figure 4 is a schematic diagram of the positions of a steel coil and a coil carrier according to another example embodiment of the present application;

[0023] Figure 5 is a schematic diagram of a steel coil centering recognition system according to an example embodiment of the present application;

[0024] Figure 6 is Figure 5 is a schematic diagram of a centering recognition process of the steel coil centering recognition system according to the example embodiment shown in the figure;

[0025] Figure 7 is a block diagram of a steel coil centering recognition device according to an example embodiment of the present application;

[0026] Figure 8 is a structural schematic diagram of an electronic device according to an example embodiment of the present application. DETAILED DESCRIPTION

[0027] The advantages and effects of the present application can be easily understood by those skilled in the art from the description of the specific embodiments. The present application can also be implemented or applied in other different specific embodiments, and various modifications or changes can be made to the details in the description without departing from the spirit of the present application. It should be noted that the following embodiments and features in the embodiments can be combined with each other without conflict.

[0028] It should be noted that the diagrams provided in the following embodiments only schematically illustrate the basic concept of the present application, and only the components related to the present application are shown in the diagrams, not the number, shape and size of the components when actually implemented. The actual implementation of each component may be a random change, and the component layout pattern may be more complex.

[0029] It should be noted that in the present application, "first", "second", etc. are only for distinguishing similar objects, not for limiting the order or sequence of similar objects. The described "includes", "has", etc. variations indicate that the subject covered by the word includes the scope of the examples shown by the word, and is not exclusive.

[0030] It can be understood that the various numbers, step numbers, etc. in the present application are distinguished for convenience of description, and do not limit the scope of the present application. The size of the present application number does not mean the order of execution, and the execution order of each process should be determined by its function and internal logic.

[0031] In the following description, a large number of details are discussed to provide a more thorough explanation of the embodiments of the present application, however, it is obvious to those skilled in the art that the embodiments of the present application can be implemented without these specific details, and in other embodiments, well-known structures and devices are shown in the form of block diagrams rather than in detail, to avoid making the embodiments of the present application difficult to understand.

[0032] It should be noted that in the steel coil production site, after the coil car takes down the steel coil from the coiler, the steel coil needs to be transported to the unloading area and unloaded. When the coil car carrying the steel coil arrives at the unloading area to start unloading, the lifting platform of the coil car lifts the steel coil and unloads the steel coil. Since the width of the lifting platform is less than the width of the steel coil, in order to avoid tilting and causing safety accidents such as rolling over when the lifting platform lifts the steel coil, the steel coil needs to be centered and identified, that is, the steel coil centering identification.

[0033] Embodiments of the present application respectively provide a steel coil centering recognition method, a steel coil centering recognition device, an electronic device, a computer readable storage medium and a computer program product, which will be described in detail below.

[0034] Please refer to Figure 1 , Figure 1 is a schematic diagram of an implementation environment of a steel coil centering recognition method according to an example embodiment of the present application.

[0035] As Figure 1 shown, the implementation environment can include an image acquisition device 101, a computer device 102 and an alarm device 103, wherein the image acquisition device 101 can be at least one of an optical camera, a digital camera, a camera, a video camera, a scanner, etc., the computer device 102 can be at least one of a desktop graphic processing unit (GPU) computer, a GPU computing cluster, a neural network computer, etc., and the alarm device 103 can be at least one of a sound alarm, a voice alarm, a light alarm and a sound and light alarm. The computer device 102 can be used to automatically process steel coil production images to obtain position information of the steel coil and position information of the coil transport vehicle, so as to determine the centering state of the steel coil. The image acquisition device 101 is used to acquire steel coil production images and provide them to the computer device 102 for processing. When the computer device 102 determines that the centering state of the steel coil is abnormal, it controls the alarm device 103 to alarm.

[0036] For example, the computer device 102 acquires a current steel coil production image acquired by the image acquisition device 101, wherein the current steel coil production image includes a steel coil and a coil transport vehicle carrying the steel coil; performs target detection on the current steel coil production image to obtain a detection result, wherein the detection result includes current position information of the steel coil and current position information of the coil transport vehicle; determines a separation state of the steel coil and the coil transport vehicle based on the current position information of the steel coil and the current position information of the coil transport vehicle, and if the separation state of the steel coil and the coil transport vehicle is separated, determines the centering state of the steel coil according to the current position information of the steel coil; and if the centering state of the steel coil is abnormal, alarms. It can be seen that the technical solution of the embodiments of the present application can automatically recognize the centering of the steel coil, avoid human factor interference, improve the accuracy and stability of the steel coil centering recognition, alarm in time when the centering is abnormal, and thus avoid safety accidents. At the same time, it can also replace manual steel coil centering recognition, effectively improve the efficiency of steel coil centering recognition, and reduce the workload of relevant personnel.

[0037] It should be noted that the steel coil centering recognition method provided by the embodiments of the present application is generally executed by the computer device 102, and the corresponding steel coil centering recognition device is generally arranged in the computer device 102.

[0038] Referring to Figure 2 , Figure 2 is a flowchart of a steel coil centering recognition method shown in an exemplary embodiment of the present application. The method can be applied to Figure 1 the implementation environment shown and be specifically executed by the computer device 102 in the implementation environment. It should be understood that the method can also be applied to other exemplary implementation environments and be specifically executed by devices in other implementation environments, and the present embodiment does not limit the implementation environment to which the method is applied.

[0039] As Figure 2 shown, in an exemplary embodiment, the steel coil centering recognition method at least includes steps S210 to S240, which are described in detail as follows:

[0040] Step S210, obtaining a current steel coil production image including a steel coil and a coil carrier carrying the steel coil.

[0041] In an embodiment of the present application, the steel coil production image refers to an image collected by an image collection device during the process in which a coil carrier delivers a produced steel coil to an unloading area such as a baling station or a temporary storage area and unloads the steel coil in a steel coil production site. Correspondingly, the current steel coil production image refers to a steel coil production image collected by the image collection device at the current time, wherein the current steel coil production image includes a sub-image of the steel coil and a sub-image of the coil carrier carrying the steel coil, and the coil carrier is also called a steel coil trolley, a coil trolley or a coil carrier. The image collection device can be a camera device or a scanning device, etc., which is not limited here.

[0042] Taking the camera device as an example, since there are multiple steel coil production lines in the steel coil production site, a camera device can be configured on each steel coil production line before collecting the steel coil production image, and each camera device is matched with the coil carrier in the steel coil production line where the camera device is located.

[0043] Step S220, performing target detection on the current steel coil production image to obtain a detection result.

[0044] In an embodiment of the present application, the detection result comprises current position information of the steel coil and current position information of the coil carrier. The current steel coil production image can be preprocessed first, including at least one of cropping, color enhancement, denoising, size adjustment, etc., and then the target detection is performed on the preprocessed current steel coil production image. There are various ways of target detection. The current steel coil production image can be subjected to feature extraction to obtain a feature map of the current steel coil production image, the feature map of the current steel coil production image is matched according to a preset steel coil feature map to identify the current position information of the steel coil in the current steel coil production image, and the feature map of the current steel coil production image is matched according to a preset coil carrier feature map to identify the current position information of the coil carrier in the current steel coil production image.

[0045] In addition, the target detection model can also be used to perform target detection on the current steel coil production image, which is not limited here.

[0046] In an embodiment of the present application, before step S220, the method comprises: acquiring a plurality of historical steel coil production images, each historical steel coil production image comprising a historical steel coil and a historical coil carrier carrying the historical steel coil; performing data labeling on the historical steel coil and the historical coil carrier in each historical steel coil production image to obtain a sample data set; training the initial target detection model using the sample data set to obtain a trained target detection model, so as to perform target detection on the current steel coil production image by using the trained target detection model.

[0047] In this embodiment, a plurality of historical steel coil production images can be acquired by using an image acquisition device, each historical steel coil production image comprising a historical steel coil and a historical coil carrier carrying the historical steel coil, each historical steel coil production image is preprocessed and data labeled to obtain a sample data set, and the initial target detection model is trained by using the sample data set.

[0048] Illustratively, the effective information of the sample data set used for training can include image basic attributes and annotation information, wherein the image basic attributes can include: filename (file name), width (width), height (height), and depth (image depth), and the annotation information can include target object corresponding target frame data and type, and the target frame data includes a pair of opposite diagonal corner coordinates of the target frame, for example: the annotation information can be [x1, y1, x2, y2, class], wherein x1 is the left upper corner horizontal coordinate of the target object corresponding target frame in the historical steel coil production image, y1 is the left upper corner vertical coordinate of the target object corresponding target frame in the historical steel coil production image, x2 is the right lower corner horizontal coordinate of the target object corresponding target frame in the historical steel coil production image, y2 is the right lower corner vertical coordinate of the target object corresponding target frame in the historical steel coil production image, and class is the category of the target object, i.e. steel coil or coil carrier, and the target object includes steel coil and coil carrier. It should be noted that the corner coordinates in the annotation information can also be another pair of opposite diagonal corner coordinates of the target frame, or the corner coordinates of the four corners of the target frame, which are not limited here. The initial target detection model obtains the features of the target object in the target frame in the deep learning sample data set and the category thereof, and finally obtains the trained target detection model, wherein the target detection model includes at least one of a recognition model and a classification model.

[0049] In addition, the sample data set can also be divided into three parts: training set, test set and validation set, the training set is used to train the target detection model, the validation set is used to adjust the hyperparameters of the target detection model and preliminarily evaluate the ability of the target detection model, and the test set is used to evaluate the final generalization ability of the target detection model. The generalization ability of the target detection model can be improved, and a target detection model with higher accuracy can be obtained.

[0050] The current steel coil production image is input into the trained target detection model, so that the trained target detection model outputs the detection frame data and the category corresponding to the target object, wherein the detection frame data can include a pair of opposite diagonal corner coordinates of the detection frame. If the category of a target object is a steel coil, the detection frame data corresponding to the target object is the current position information of the steel coil; if the category of a target object is a coil carrier, the detection frame data corresponding to the target object is the current position information of the coil carrier. By using the target detection model to detect the steel coil production image, the efficiency of target detection can be improved, and the workload of target detection can be effectively reduced.

[0051] In step S230, the separation state of the steel coil and the coil carrier is determined based on the current position information of the steel coil and the current position information of the coil carrier in the detection result, and if the separation state of the steel coil and the coil carrier is separated, the centering state of the steel coil is determined according to the current position information of the steel coil.

[0052] In an embodiment of the present application, when the coil carrying vehicle carrying the steel coil reaches the coil unloading area, the lifting platform of the coil carrying vehicle lifts the steel coil and unloads the steel coil, completing the unloading of the steel coil. When the lifting platform lifts the steel coil, the steel coil is separated from the coil carrying vehicle, and if the center line of the steel coil is not aligned with the center line of the lifting platform, the steel coil may be tilted during lifting, causing safety accidents such as coil overturning. Therefore, the relative position of the steel coil and the coil carrying vehicle at the current time can be determined based on the current position information of the steel coil and the current position information of the coil carrying vehicle, and the relative position of the steel coil and the coil carrying vehicle at the current time is compared with the relative position of the steel coil and the coil carrying vehicle at the previous time. If the relative position of the steel coil and the coil carrying vehicle gradually moves away in height, it indicates that the steel coil and the coil carrying vehicle are separated, i.e., the separation state of the steel coil and the coil carrying vehicle is separated. If the relative position of the steel coil and the coil carrying vehicle does not change in height, it indicates that the steel coil and the coil carrying vehicle are not separated, i.e., the separation state of the steel coil and the coil carrying vehicle is not separated. When it is judged that the steel coil and the coil carrying vehicle are separated, in order to avoid the occurrence of safety accidents, it is necessary to further judge whether the steel coil is centered. There are various methods to judge whether the steel coil is centered. Generally, the lifting platform is located at the center of the coil carrying vehicle, the center line position of the steel coil can be determined according to the current position information of the steel coil, the center line position of the coil carrying vehicle can be determined according to the current position information of the coil carrying vehicle, and the center line position of the steel coil and the center line position of the coil carrying vehicle are compared to determine whether the steel coil is centered, i.e., to determine the centering state of the steel coil. Alternatively, the center line position when the lifting platform is lifted can be calculated in advance and used as a preset center line position. The center line position of the steel coil can be determined according to the current position information of the steel coil, and the center line position of the steel coil and the preset center line position are compared to determine whether the steel coil is centered.

[0053] It should be understood that the relative position of the steel coil and the coil carrying vehicle at the previous time can be obtained by processing the steel coil production image at the previous time according to the processing method of the current steel coil production image. For details, please refer to the content described in the embodiments, which will not be repeated here.

[0054] In an embodiment of the present application, before step S230, the method comprises: comparing the current position information of the steel coil with the preset region of interest position information of the preset region of interest to determine the relative position relationship between the steel coil and the preset region of interest; comparing the current position information of the coil carrying vehicle with the preset region of interest position information to determine the relative position relationship between the coil carrying vehicle and the preset region of interest; and if the steel coil and the coil carrying vehicle are both located in the preset region of interest, determining the separation state of the steel coil and the coil carrying vehicle based on the current position information of the steel coil and the current position information of the coil carrying vehicle.

[0055] In this embodiment, since it takes a certain length of time for the coil transport vehicle to transport the finished steel coil to the coil unloading area, if the separation of the steel coil and the coil transport vehicle is determined at each moment during the transportation, the calculation amount is large, and unnecessary resource waste is easily caused. Therefore, the separation of the steel coil and the coil transport vehicle can be determined when the coil transport vehicle reaches a certain area, which can effectively reduce the calculation amount and save resources. The field of view of each camera device can be pre-set with a region of interest as a preset region of interest, and the position information of the preset region of interest in the field of view of the camera device can be determined as preset region position information. Illustratively, the preset region of interest can be a rectangle, and correspondingly, the preset region position information can be at least a pair of opposite diagonal corner coordinates of the rectangle corresponding to the preset region of interest. The preset region of interest can also be other shapes, which are not limited here. The current position information of the steel coil and the current position information of the coil transport vehicle are compared with the preset region position information respectively, to determine whether the steel coil and the coil transport vehicle are both located in the preset region of interest. If at least one of the steel coil and the coil transport vehicle is not in the preset region of interest, subsequent determination is not performed, and the next steel coil production image is acquired at the next moment, and the steel coil centering recognition is performed on the next steel coil production image; if both the steel coil and the coil transport vehicle are in the preset region of interest, it is further determined whether the steel coil and the coil transport vehicle are separated.

[0056] In an embodiment of the present application, the separation state of the steel coil and the coil carrier is determined based on the current position information of the steel coil and the current position information of the coil carrier, including: comparing the current top-left longitudinal coordinate of the coil carrier corresponding detection box in the current position information of the coil carrier with the current bottom-right longitudinal coordinate of the steel coil corresponding detection box in the current position information of the steel coil to obtain a first comparison result; comparing the current bottom-right horizontal coordinate of the coil carrier corresponding detection box in the current position information of the coil carrier with the preset center line horizontal coordinate to obtain a second comparison result; calculating the difference between the initial bottom-right longitudinal coordinate of the steel coil corresponding detection box and the current bottom-right longitudinal coordinate of the steel coil corresponding detection box as a first difference value, wherein the initial bottom-right longitudinal coordinate of the steel coil corresponding detection box is obtained based on target detection on the initial steel coil production image; calculating the difference between the current top-left longitudinal coordinate of the coil carrier corresponding detection box and the current bottom-right longitudinal coordinate of the steel coil corresponding detection box as a second difference value; if the first comparison result meets the first preset condition, and the second comparison result meets the second preset condition, and the first difference value meets the first preset difference value, and the second difference value meets the second preset difference value, the separation state of the steel coil and the coil carrier is separated; if the first comparison result does not meet the first preset condition, and / or the second comparison result does not meet the second preset condition, and / or the first difference value does not meet the first preset difference value, and / or the second difference value does not meet the second preset difference value, the separation state of the steel coil and the coil carrier is not separated.

[0057] Please refer to Figure 3 and Figure 4 , Figure 3 is a schematic diagram of the position of the steel coil and the coil carrier shown in an embodiment of the present application, Figure 4 is a schematic diagram of the position of the steel coil and the coil carrier shown in another embodiment of the present application. As shown in Figure 3 and Figure 4 , a coordinate system is established with the top-left corner of the current steel coil production image as the origin, and target detection is performed on the current production image to obtain the current position information of the steel coil and the current position information of the coil carrier, wherein the current position information of the steel coil includes the current top-left horizontal coordinate x 1coil , the current top-left longitudinal coordinate y 1coil , the current bottom-right horizontal coordinate x 2coil and the current bottom-right longitudinal coordinate y 2coil of the steel coil corresponding detection box (steel coil detection box), and the current position information of the coil carrier includes the current top-left horizontal coordinate x 1car , the current top-left longitudinal coordinate y 1car , the current bottom-right horizontal coordinate x 2car and the current bottom-right longitudinal coordinate y2car Based on the current position information of the steel coil and the current position information of the coil transport vehicle, if the following conditions are met, it is determined that the separation state of the steel coil and the coil transport vehicle is separated:

[0058] y 1car >y 2coil and x 2car >x base and y 2difference >m and distance>n

[0059] wherein y 1car is the current top-left longitudinal coordinate of the detection frame corresponding to the coil transport vehicle, y 2coil is the current bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil, y 1car >y 2coil is a first preset condition, x 2car is the current bottom-right horizontal coordinate of the detection frame corresponding to the coil transport vehicle, x base is a preset center line horizontal coordinate, x 2car >x base is a second preset condition, y 2difference is the initial bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil, y 2coil_初 is the initial bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil, y 2coil is the difference between y 1car and y 2coil , i.e. the first difference, distance is the difference between y 1car and y 2coil , i.e. the second difference, m is a first preset difference, and n is a second preset difference. Illustratively, the first preset difference can be 1 pixel, or other numerical values, and the second preset difference can be 3 pixels, or other numerical values, which are not limited here.

[0060] It should be understood that the initial bottom-right longitudinal coordinate of the detection frame corresponding to the steel coil can be obtained by target detection on the initial steel coil production image, i.e. the initial steel coil production image, in the manner of target detection on the current steel coil production image. For details, please refer to the contents recorded in each embodiment, which will not be repeated here.

[0061] If the above conditions are not met, it is determined that the separation state of the steel coil and the coil transport vehicle is not separated.

[0062] In one embodiment of the present application, the centering state of the steel coil is determined according to the current position information of the steel coil, including: calculating the current center horizontal coordinate of the steel coil corresponding detection frame according to the current left upper corner horizontal coordinate of the steel coil corresponding detection frame and the current right lower corner horizontal coordinate of the steel coil corresponding detection frame, the current position information of the steel coil including the current left upper corner horizontal coordinate of the steel coil corresponding detection frame and the current right lower corner horizontal coordinate of the steel coil corresponding detection frame; calculating the difference between the current center horizontal coordinate of the steel coil corresponding detection frame and the preset center line horizontal coordinate as the initial offset of the steel coil, and determining the actual offset of the steel coil according to the initial offset of the steel coil and the preset scale; if the actual offset of the steel coil is less than or equal to the preset offset, the centering state of the steel coil is normal; if the actual offset of the steel coil is greater than the preset offset, the centering state of the steel coil is abnormal.

[0063] In this embodiment, the current center horizontal coordinate of the steel coil corresponding detection frame is calculated as follows:

[0064] x = (x 1coil +x 2coil ) / 2 Formula (1),

[0065] Wherein, x is the current center horizontal coordinate of the steel coil corresponding detection frame, x 1coil is the current left upper corner horizontal coordinate of the steel coil corresponding detection frame, and x 2coil is the current right lower corner horizontal coordinate of the steel coil corresponding detection frame.

[0066] The initial offset of the steel coil is calculated as follows:

[0067] offset = |x-x base | Formula (2),

[0068] Wherein, offset is the initial offset of the steel coil, x is the current center horizontal coordinate of the steel coil corresponding detection frame, and x base is the preset center line horizontal coordinate.

[0069] The initial offset of the steel coil is the pixel offset corresponding to the steel coil in the current steel coil production image, which needs to be converted into the actual offset of the steel coil, and the actual offset of the steel coil is calculated as follows:

[0070] offset scaled = offset x scale Formula (3),

[0071] Wherein, offset scaled is the actual offset of the steel coil, offset is the initial offset of the steel coil, and scale is the preset scale.

[0072] After obtaining the actual offset of the steel coil, the actual offset of the steel coil is compared with the preset offset. If the actual offset of the steel coil is less than or equal to the preset offset, the steel coil is centered normally, that is, the centering state of the steel coil is normal centering. Otherwise, the steel coil is centered abnormally, that is, the centering state of the steel coil is abnormal centering. Illustratively, the preset offset can be 50 mm or other values, which are not limited here.

[0073] In step S240, if the centering state of the steel coil is abnormal centering, an alarm is given.

[0074] In an embodiment of the present application, if the centering state of the steel coil is normal centering, the uncoiling continues; if the centering state of the steel coil is abnormal centering, an alarm is given, and an alarm signal can be sent to an alarm device to give an alarm through the alarm device. The alarm device can be at least one of a sound alarm, a voice alarm, a light alarm and a sound and light alarm, so that the staff can discover the abnormal centering of the steel coil in time, and the lifting platform is paused to rise, which can effectively avoid safety accidents when the steel coil has abnormal centering. In addition, if the centering state of the steel coil is abnormal centering, a pause rising instruction can be generated and sent to the control system of the lifting platform at the same time to pause the rising of the lifting platform.

[0075] In another embodiment of the present application, after determining the relative position relationship between the coil carrying vehicle and the preset region of interest, the method further comprises: when there are multiple steel coils and the coil carrying vehicle carrying each steel coil is located in the preset region of interest, calculating the current area of the detection frame corresponding to each steel coil according to the current corner point coordinate information of the detection frame corresponding to each steel coil. The current position information of the steel coil includes the current corner point coordinate information of the detection frame corresponding to the steel coil; if the current area of the detection frame corresponding to a steel coil meets the preset area interval corresponding to the camera near point, the steel coil is taken as a target steel coil, and the coil carrying vehicle carrying the steel coil is taken as a target coil carrying vehicle. The separation state of the target steel coil and the target coil carrying vehicle is determined based on the current position information of the target steel coil and the current position information of the target coil carrying vehicle.

[0076] In this embodiment, since there are multiple steel coil production lines in the steel coil production site, multiple groups of steel coil-pallet trucks may exist in the current steel coil production image collected, and each group of steel coil-pallet truck includes a steel coil and a pallet truck carrying the steel coil. Correspondingly, if multiple steel coils and the pallet trucks carrying each steel coil exist in the preset region of interest, it is necessary to determine the target steel coil that needs to be identified from the multiple steel coils in the preset region of interest. The current area of each steel coil corresponding detection box can be compared with the preset camera near point corresponding area interval. If the current area of a steel coil corresponding detection box is within the preset camera near point corresponding area interval, it indicates that the steel coil is located at the preset camera near point of the image collection device, and the steel coil can be taken as the target steel coil, and the pallet truck carrying the steel coil can be taken as the target pallet truck, and then the separation state of the target steel coil and the target pallet truck is judged.

[0077] The current area of each steel coil corresponding detection box can be calculated according to the current corner point coordinate information of each steel coil corresponding detection box. For example, the current corner point coordinate information of a steel coil corresponding detection box includes the current left upper corner horizontal coordinate, the current left upper corner vertical coordinate, the current right lower corner horizontal coordinate, and the current right lower corner vertical coordinate. The calculation method of the current area of the steel coil corresponding detection box is as follows:

[0078] area = |x 2coil -x 1coil | x | y 2coil -y 1coil | Formula (4),

[0079] wherein, area is the current area of the steel coil corresponding detection box, x 1coil is the current left upper corner horizontal coordinate of the steel coil corresponding detection box, y 1coil is the current left upper corner vertical coordinate of the steel coil corresponding detection box, x 2coil is the current right lower corner horizontal coordinate of the steel coil corresponding detection box, and y 2coil is the current right lower corner vertical coordinate of the steel coil corresponding detection box.

[0080] The preset camera near point corresponding area interval is: [area coil1 (1-padding), area coil1 (1+padding)], wherein, area coil1 is the area size of the preset camera near point on the image, and padding is the preset allowable deviation.

[0081] In another embodiment of the present application, if the current area of each coil corresponding detection frame does not satisfy the preset camera near point corresponding area interval, the method comprises: comparing the current area of each coil corresponding detection frame with the preset camera far point corresponding area interval; if the current area of a coil corresponding detection frame satisfies the preset camera far point corresponding area interval, taking the coil as a target coil and taking the coil carrying vehicle carrying the coil as a target coil carrying vehicle, and determining the separation state of the target coil and the target coil carrying vehicle based on the current position information of the target coil and the current position information of the target coil carrying vehicle.

[0082] In this embodiment, if the current area of each coil corresponding detection frame in the preset region of interest is not in the preset camera near point corresponding area interval, it indicates that no coil is located at the preset camera near point of the image acquisition device. Based on this, the current area of each coil corresponding detection frame can be compared with the preset camera far point corresponding area interval, if the current area of a coil corresponding detection frame is in the preset camera far point corresponding area interval, it indicates that the coil is located at the preset camera far point of the image acquisition device, the coil can be taken as a target coil, and the coil carrying vehicle carrying the coil can be taken as a target coil carrying vehicle, and then the separation state of the target coil and the target coil carrying vehicle is determined.

[0083] The preset camera far point corresponding area interval is: [area coil2 (1-padding), area coil2 (1+padding)], wherein area coil2 is the area size of the preset camera far point on the image, and padding is a preset allowable deviation.

[0084] In another embodiment of the present application, determining the separation state of the target steel coil and the target coil handling vehicle based on the current position information of the target steel coil and the current position information of the target coil handling vehicle comprises: comparing the current top-left corner longitudinal coordinate of the target coil handling vehicle corresponding detection box with the current bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box to obtain a first comparison result, the current position information of the target coil handling vehicle includes the current top-left corner longitudinal coordinate of the target coil handling vehicle corresponding detection box, and the current position information of the target steel coil includes the current bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box; comparing the current bottom-right corner transverse coordinate of the target coil handling vehicle corresponding detection box with a preset center line transverse coordinate to obtain a second comparison result, the current position information of the target coil handling vehicle also includes the current bottom-right corner transverse coordinate of the target coil handling vehicle corresponding detection box; calculating the difference between the initial bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box and the current bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box as a first difference, the initial bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box is obtained based on target detection on the initial steel coil production image; calculating the difference between the current top-left corner longitudinal coordinate of the target coil handling vehicle corresponding detection box and the current bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box as a second difference; if the first comparison result meets a first preset condition, and the second comparison result meets a second preset condition, and the first difference is greater than a first preset difference, and the second difference is greater than a second preset difference, then the separation state of the target steel coil and the target coil handling vehicle is separated; if the first comparison result does not meet the first preset condition, and / or the second comparison result does not meet the second preset condition, and / or the first difference is less than or equal to the first preset difference, and / or the second difference is less than or equal to the second preset difference, then the separation state of the target steel coil and the target coil handling vehicle is not separated.

[0085] In a specific embodiment of the present application, a coordinate system is established with the top-left corner of the current steel coil production image as the origin, and if the following conditions are met, it is determined that the separation state of the target steel coil and the target coil handling vehicle is separated:

[0086] y 1car_目标 >y 2coil_目标 and x 2car_目标 >x base and y 2difference_目标 >m and distance _目标 >n

[0087] wherein y 1car_目标 is the current top-left corner longitudinal coordinate of the target coil handling vehicle corresponding detection box, y 2coil_目标 is the current bottom-right corner longitudinal coordinate of the target steel coil corresponding detection box, y 1car_目标 >y 2coil_目标 is the first preset condition, x 2car_目标 is the current bottom-right corner transverse coordinate of the target coil handling vehicle corresponding detection box, x base is the preset center line transverse coordinate, and x2car_目标 >x base For the second preset condition, y 2difference_目标 is the initial lower right corner longitudinal coordinate y of the target coil corresponding detection frame 2coil_初_目标 and y 2coil_目标 , that is, the first difference distance _目标 is y 1car_目标 and y 2coil_目标 , that is, the second difference m, the first preset difference is n, the second preset difference. Illustratively, the first preset difference can be 1 pixel, or other numerical value, the second preset difference can be 3 pixels, or other numerical value, not limited here.

[0088] It should be understood that the initial lower right corner longitudinal coordinate of the target coil corresponding detection frame can be obtained by target detection on the initial coil production image, that is, the initial steel coil production image, according to the target detection on the current steel coil production image. For details, please refer to the contents recorded in each embodiment, which will not be repeated here.

[0089] If the above conditions are not met, it is determined that the separation state of the target coil and the target coil car is not separated.

[0090] In another embodiment of the present application, if the separation state of the target coil and the target coil car is separated, the method comprises: calculating the current center horizontal coordinate of the target coil corresponding detection frame according to the current upper left corner horizontal coordinate of the target coil corresponding detection frame and the current lower right corner horizontal coordinate of the target coil corresponding detection frame, the current position information of the target coil comprises the current upper left corner horizontal coordinate of the target coil corresponding detection frame and the current lower right corner horizontal coordinate of the target coil corresponding detection frame; calculating the difference between the current center horizontal coordinate of the target coil corresponding detection frame and the preset center line horizontal coordinate as the initial offset of the target coil, and determining the actual offset of the target coil according to the initial offset of the target coil and the preset proportion; if the actual offset of the target coil is less than or equal to the preset offset, the centering state of the target coil is normal; if the actual offset of the target coil is greater than the preset offset, the centering state of the target coil is abnormal.

[0091] In this embodiment, the calculation method of the current center horizontal coordinate of the target coil corresponding detection frame is as follows:

[0092] x _目标 =(x 1coil_目标 +x 2coil_目标 ) / 2 formula (5),

[0093] Wherein, x _目标 is the current center horizontal coordinate of the target coil corresponding detection frame, x 1coil_目标 is the current upper left corner horizontal coordinate of the target coil corresponding detection frame, x2coil_目标 is the current right bottom horizontal coordinate of the detection frame corresponding to the target steel coil.

[0094] The initial offset of the target steel coil is calculated as follows:

[0095] offset _目标 = |x _目标 - x base | Equation (6),

[0096] wherein offset _目标 is the initial offset of the target steel coil, x is the current center horizontal coordinate of the detection frame corresponding to the target steel coil, and x base is the preset center line horizontal coordinate.

[0097] The initial offset of the target steel coil is the pixel offset of the target steel coil in the current steel coil production image, which needs to be converted into the actual offset of the target steel coil. The actual offset of the target steel coil is calculated as follows:

[0098] offset scaled_目标 = offset _目标 x scale Equation (7),

[0099] wherein offset scaled_目标 is the actual offset of the target steel coil, offset _目标 is the initial offset of the target steel coil, and scale is a preset scale.

[0100] After obtaining the actual offset of the target steel coil, the actual offset of the target steel coil is compared with the preset offset. If the actual offset of the target steel coil is less than or equal to the preset offset, the target steel coil is centered normally, i.e., the centering state of the target steel coil is centered normally. Otherwise, the target steel coil is centered abnormally, i.e., the centering state of the target steel coil is centered abnormally. Illustratively, the preset offset can be 50 mm or other numerical values, which are not limited herein.

[0101] Please refer to Figure 5 , Figure 5 is a structural schematic diagram of a steel coil centering recognition system according to an embodiment of the present application. As shown in Figure 5As shown, the steel coil centering recognition system includes a collection module, a training module, and a determination module. The collection module is configured to collect a plurality of historical steel coil production images of a steel coil production site, and pre-process and label each historical steel coil production image to obtain a sample data set. The training module is configured to train a deep learning network through the sample data set to obtain a target detection model, which can include at least one of an identification model, a classification model, etc. The determination module is configured to perform target detection on a real-time steel coil production image through the target detection model to obtain a detection result, including real-time position information of the steel coil and real-time position information of the coil carrier. The real-time position information of the steel coil, the real-time position information of the coil carrier, and preset center line position information are used to determine whether the steel coil is separated from the coil carrier, and whether the steel coil is centered when the steel coil is separated from the coil carrier.

[0102] Referring to Figure 6 , Figure 6 is Figure 5 FIG. 1 shows a schematic diagram of a centering recognition process of the steel coil centering recognition system in the embodiment. As shown, Figure 6 The steel coil centering recognition process of the steel coil centering recognition system is as follows:

[0103] 1. A plurality of historical steel coil production images of a steel coil production site are collected by a camera device, and each historical steel coil production image is pre-processed and labeled to obtain a sample data set.

[0104] 2. A deep learning network is trained through the sample data set to obtain a target detection model, which can include at least one of an identification model, a classification model, etc.

[0105] 3. A current steel coil production image is collected by the camera device, pre-processed, and input into the target detection model to obtain a detection result, including current position information of the steel coil and current position information of the coil carrier.

[0106] 4. When the steel coil is separated from the coil carrier, the actual offset of the steel coil is calculated to determine whether the steel coil is centered abnormally.

[0107] 5. When the actual offset of the steel coil is greater than a preset offset, an alarm is given.

[0108] Figure 6For details of the procedures in the flow, please refer to the descriptions in the foregoing embodiments, which will not be repeated here. The technical solution of this embodiment collects historical steel coil production images including a steel coil and a steel coil trolley carrying the steel coil by using a camera device, pre-processes and labels the steel coil production images, inputs the labeled historical steel coil production images into a deep learning network to train the deep learning network, improves the generalization ability of the original model, obtains a target detection model with higher accuracy, and then obtains real-time position information of the steel coil and real-time position information of the coil trolley in real-time steel coil production images by using the target detection model to determine whether the steel coil is centered, which can effectively avoid human errors in manual determination and occurrence of hazards to construction safety.

[0109] Please refer to Figure 7 , Figure 7 is a block diagram of a steel coil centering recognition device according to an example embodiment of the present application. The device can be applied to the implementation environment shown in Figure 1 , and is specifically configured in the computer device 102. The device can also be applied to other example implementation environments, and is specifically configured in other devices. The implementation environment to which the device is applied is not limited in the present embodiment.

[0110] As shown in Figure 7 , the example steel coil centering recognition device includes:

[0111] The acquisition module 710 is configured to acquire a current steel coil production image, the current steel coil production image including a steel coil and a coil trolley carrying the steel coil. The detection module 720 is configured to perform target detection on the current steel coil production image to obtain a detection result, the detection result including current position information of the steel coil and current position information of the coil trolley. The determination module 730 is configured to determine a separation state of the steel coil and the coil trolley based on the current position information of the steel coil and the current position information of the coil trolley, and determine a centering state of the steel coil according to the current position information of the steel coil if the separation state of the steel coil and the coil trolley is separated. The alarm module 740 is configured to alarm if the centering state of the steel coil is abnormal.

[0112] It should be noted that the steel coil centering recognition device provided in the above embodiments and the steel coil centering recognition method provided in the above embodiments belong to the same concept, and the specific manner in which each module and unit performs operations has been described in detail in the method embodiments, which will not be repeated here. In actual application, the functions of the above embodiments can be distributed to different functional modules to complete all or part of the functions described above, i.e., the internal structure of the device is divided into different functional modules to complete all or part of the functions described above, and this will not be limited here.

[0113] The embodiment also provides an electronic device, comprising: one or more processors; a storage device for storing one or more programs, which, when executed by the one or more processors, cause the electronic device to implement the steel coil centering identification method provided in each of the above embodiments.

[0114] Referring to Figure 8 , Figure 8 is a structural schematic diagram of an electronic device according to an example embodiment of the present application. It should be noted that Figure 8 The electronic device 800 shown is only an example, and should not impose any limitation on the functions and use range of the embodiments of the present application.

[0115] As Figure 8 shown, the electronic device 800 includes a processor 801, a memory 802 and a communication bus 803; the communication bus 803 is used to connect the processor 801 and the memory 802; the processor 801 is used to execute the computer program stored in the memory 802, so as to implement the method of one or more of the above embodiments.

[0116] The embodiment also provides a computer readable storage medium, which stores a computer program, and the computer program is executed by a processor of a computer, so that the computer executes the steel coil centering identification method as described above. The computer readable storage medium can be included in the electronic device described in the above embodiments, or can exist separately and not be assembled into the electronic device.

[0117] The embodiment also provides a computer program product or computer program, which includes computer instructions stored in a computer readable storage medium. The processor of a computer device reads the computer instructions from the computer readable storage medium, and the processor executes the computer instructions, so that the computer device executes the steel coil centering identification method provided in each of the above embodiments.

[0118] The electronic device provided by the embodiment includes a processor, a memory, a transceiver and a communication interface, the memory and the communication interface are connected with the processor and the transceiver and complete communication between each other, the memory is used to store a computer program, the communication interface is used to communicate, and the processor and the transceiver are used to run the computer program, so that the electronic device executes each step of the above method.

[0119] In the embodiment, the memory can include a random access memory (RAM) and can also include a non-volatile memory, for example, at least one disk memory.

[0120] The processor described above can be a general processor, including a central processing unit (CPU), a network processor (NP), etc.; can also be a digital signal processor (DSP), an application specific integrated circuit (ASIC), a field-programmable gate array (FPGA) or other programmable logic device, a discrete gate or transistor logic device, a discrete hardware component.

[0121] The computer readable storage medium in the embodiment can be understood by those skilled in the art that all or part of the steps of the above-mentioned method embodiments can be completed by a computer program related hardware. The foregoing computer program can be stored in a computer readable storage medium. When the program is executed, the steps of the above-mentioned method embodiments are executed; and the foregoing storage medium includes ROM (read only memory), RAM (random access memory), magnetic disk or optical disk and various storage media that can store program codes.

[0122] The above-mentioned embodiments only exemplarily illustrate the principles and effects of the present application, and are not used to limit the present application. Any person skilled in the art can modify or change the above-mentioned embodiments without departing from the spirit and scope of the present application. Therefore, all equivalent modifications or changes made by those skilled in the art without departing from the spirit and technical idea of the present application should be covered by the claims of the present application.

Claims

1. A steel coil centering recognition method characterized by, The method comprises: acquiring a current steel coil production image, the current steel coil production image comprising a steel coil and a coil carrier carrying the steel coil; performing target detection on the current steel coil production image to obtain a detection result, the detection result comprising current position information of the steel coil and current position information of the coil carrier; determining a separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier, and if the separation state of the steel coil and the coil carrier is separated, determining a centering state of the steel coil based on the current position information of the steel coil; the determination of the separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier comprises: comparing a current top-left vertical coordinate of a detection box corresponding to the coil carrier with a current bottom-right vertical coordinate of a detection box corresponding to the steel coil to obtain a first comparison result, the current position information of the coil carrier comprising the current top-left vertical coordinate of the detection box corresponding to the coil carrier, and the current position information of the steel coil comprising the current bottom-right vertical coordinate of the detection box corresponding to the steel coil; comparing a current bottom-right horizontal coordinate of the detection box corresponding to the coil carrier with a preset center line horizontal coordinate to obtain a second comparison result, the current position information of the coil carrier further comprising the current bottom-right horizontal coordinate of the detection box corresponding to the coil carrier; calculating a difference between an initial bottom-right vertical coordinate of the detection box corresponding to the steel coil and the current bottom-right vertical coordinate of the detection box corresponding to the steel coil as a first difference, the initial bottom-right vertical coordinate of the detection box corresponding to the steel coil being obtained based on target detection on an initial steel coil production image; calculating a difference between the current top-left vertical coordinate of the detection box corresponding to the coil carrier and the current bottom-right vertical coordinate of the detection box corresponding to the steel coil as a second difference; if the first comparison result satisfies a first preset condition, and the second comparison result satisfies a second preset condition, and the first difference is greater than a first preset difference, and the second difference is greater than a second preset difference, the separation state of the steel coil and the coil carrier is separated; if the first comparison result does not satisfy the first preset condition, and / or the second comparison result does not satisfy the second preset condition, and / or the first difference is less than or equal to the first preset difference, and / or the second difference is less than or equal to the second preset difference, the separation state of the steel coil and the coil carrier is not separated; if the centering state of the steel coil is abnormal, performing an alarm.

2. The coil centering recognition method of claim 1, wherein Before the determination of the separation state of the steel coil and the coil carrier based on the current position information of the steel coil and the current position information of the coil carrier, the method comprises: comparing the current position information of the steel coil with preset region of interest position information of a preset region of interest to determine a relative position relationship between the steel coil and the preset region of interest; comparing the current position information of the coil carrier with the preset region of interest position information to determine a relative position relationship between the coil carrier and the preset region of interest; If the steel coil and the coil carrier are both located in the preset region of interest, a separation state of the steel coil and the coil carrier is determined based on current position information of the steel coil and current position information of the coil carrier.

3. The coil centering recognition method according to any one of claims 1 or 2, characterized in that The centering state of the steel coil is determined according to the current position information of the steel coil, including: A current center horizontal coordinate of the steel coil corresponding detection frame is calculated according to a current upper left horizontal coordinate of the steel coil corresponding detection frame and a current lower right horizontal coordinate of the steel coil corresponding detection frame, and the current position information of the steel coil includes the current upper left horizontal coordinate of the steel coil corresponding detection frame and the current lower right horizontal coordinate of the steel coil corresponding detection frame; A difference value between the current center horizontal coordinate of the steel coil corresponding detection frame and a preset center line horizontal coordinate is calculated as an initial offset of the steel coil, and an actual offset of the steel coil is determined according to the initial offset of the steel coil and a preset proportion; If the actual offset of the steel coil is less than or equal to a preset offset, the centering state of the steel coil is normal centering; If the actual offset of the steel coil is greater than the preset offset, the centering state of the steel coil is abnormal centering.

4. The coil centering recognition method of claim 2, wherein After determining the relative position relationship between the coil carrier and the preset region of interest, the method further includes: When there are multiple steel coils and the coil carrier carrying each steel coil are all located in the preset region of interest, a current area of each steel coil corresponding detection frame is calculated according to current corner point coordinate information of each steel coil corresponding detection frame, and the current position information of the steel coil includes the current corner point coordinate information of the steel coil corresponding detection frame; If the current area of a steel coil corresponding detection frame meets a preset camera near point corresponding area interval, the steel coil is taken as a target steel coil, and the coil carrier carrying the steel coil is taken as a target coil carrier, and a separation state of the target steel coil and the target coil carrier is determined based on current position information of the target steel coil and current position information of the target coil carrier.

5. The coil centering recognition method of claim 4, wherein If the current area of all steel coil corresponding detection frames does not meet the preset camera near point corresponding area interval, the method includes: The current area of each steel coil corresponding detection frame is compared with a preset camera far point corresponding area interval; If the current area of a steel coil corresponding detection frame meets the preset camera far point corresponding area interval, the steel coil is taken as a target steel coil, and the coil carrier carrying the steel coil is taken as a target coil carrier, and a separation state of the target steel coil and the target coil carrier is determined based on current position information of the target steel coil and current position information of the target coil carrier.

6. The coil centering recognition method according to any one of claims 4 or 5, characterized in that, The separation state of the target steel coil and the target coil carrier is determined based on the current position information of the target steel coil and the current position information of the target coil carrier, including: A first comparison result is obtained by comparing a current upper left vertical coordinate of the target coil carrier corresponding detection frame with a current lower right vertical coordinate of the target steel coil corresponding detection frame, and the current position information of the target coil carrier includes the current upper left vertical coordinate of the target coil carrier corresponding detection frame, and the current position information of the target steel coil includes the current lower right vertical coordinate of the target steel coil corresponding detection frame; If the first comparison result is greater than a preset first comparison threshold, the target steel coil is determined to be separated from the target coil carrier. comparing the current right-bottom horizontal coordinate of the detection frame corresponding to the target coil carrier with the preset center line horizontal coordinate to obtain a second comparison result, and the current position information of the target coil carrier further includes the current right-bottom horizontal coordinate of the detection frame corresponding to the target coil carrier; calculating a difference between the initial right-bottom vertical coordinate of the detection frame corresponding to the target coil and the current right-bottom vertical coordinate of the detection frame corresponding to the target coil as a first difference, and the initial right-bottom vertical coordinate of the detection frame corresponding to the target coil is obtained based on target detection on an initial coil production image; calculating a difference between the current left-top vertical coordinate of the detection frame corresponding to the target coil carrier and the current right-bottom vertical coordinate of the detection frame corresponding to the target coil as a second difference; if the first comparison result satisfies a first preset condition, and the second comparison result satisfies a second preset condition, and the first difference is greater than a first preset difference, and the second difference is greater than a second preset difference, then the separation state of the target coil and the target coil carrier is separated; if the first comparison result does not satisfy the first preset condition, and / or the second comparison result does not satisfy the second preset condition, and / or the first difference is less than or equal to the first preset difference, and / or the second difference is less than or equal to the second preset difference, then the separation state of the target coil and the target coil carrier is not separated.

7. The coil centering recognition method of claim 6, wherein if the separation state of the target coil and the target coil carrier is separated, the method comprises: calculating a current center horizontal coordinate of the detection frame corresponding to the target coil according to the current left-top horizontal coordinate of the detection frame corresponding to the target coil and the current right-bottom horizontal coordinate of the detection frame corresponding to the target coil, and the current position information of the target coil includes the current left-top horizontal coordinate of the detection frame corresponding to the target coil and the current right-bottom horizontal coordinate of the detection frame corresponding to the target coil; calculating a difference between the current center horizontal coordinate of the detection frame corresponding to the target coil and the preset center line horizontal coordinate as an initial offset of the target coil, and determining an actual offset of the target coil according to the initial offset of the target coil and a preset proportion; if the actual offset of the target coil is less than or equal to a preset offset, then the centering state of the target coil is normal; if the actual offset of the target coil is greater than the preset offset, then the centering state of the target coil is abnormal.

8. The coil centering recognition method according to any one of claims 1 or 2, characterized in that, Before target detection on the current coil production image, the method comprises: obtaining a plurality of historical coil production images, each historical coil production image including a historical coil and a historical coil carrier carrying the historical coil; performing data labeling on the historical coil and the historical coil carrier in each historical coil production image to obtain a sample data set; training an initial target detection model using the sample data set to obtain a trained target detection model, so as to perform target detection on the current coil production image through the trained target detection model.

9. A coil centering recognition device characterized by comprising: The device comprises: The acquisition module is configured to acquire a current steel coil production image, the current steel coil production image including a steel coil and a coil carrying vehicle carrying the steel coil; The detection module is configured to perform target detection on the current steel coil production image to obtain a detection result, the detection result including current position information of the steel coil and current position information of the coil carrying vehicle; The determination module is configured to determine a separation state of the steel coil and the coil carrying vehicle based on the current position information of the steel coil and the current position information of the coil carrying vehicle, and determine a centering state of the steel coil according to the current position information of the steel coil if the separation state of the steel coil and the coil carrying vehicle is separated. The determination of the separation state of the steel coil and the coil carrying vehicle based on the current position information of the steel coil and the current position information of the coil carrying vehicle includes: comparing a current top-left vertical coordinate of a detection frame corresponding to the coil carrying vehicle with a current bottom-right vertical coordinate of a detection frame corresponding to the steel coil to obtain a first comparison result, the current position information of the coil carrying vehicle including the current top-left vertical coordinate of the detection frame corresponding to the coil carrying vehicle, and the current position information of the steel coil including the current bottom-right vertical coordinate of the detection frame corresponding to the steel coil; comparing a current bottom-right horizontal coordinate of the detection frame corresponding to the coil carrying vehicle with a preset center line horizontal coordinate to obtain a second comparison result, the current position information of the coil carrying vehicle further including the current bottom-right horizontal coordinate of the detection frame corresponding to the coil carrying vehicle; calculating a difference between an initial bottom-right vertical coordinate of the detection frame corresponding to the steel coil and the current bottom-right vertical coordinate of the detection frame corresponding to the steel coil as a first difference, the initial bottom-right vertical coordinate of the detection frame corresponding to the steel coil being obtained based on target detection on an initial steel coil production image; calculating a difference between the current top-left vertical coordinate of the detection frame corresponding to the coil carrying vehicle and the current bottom-right vertical coordinate of the detection frame corresponding to the steel coil as a second difference; if the first comparison result satisfies a first preset condition, and the second comparison result satisfies a second preset condition, and the first difference is greater than a first preset difference, and the second difference is greater than a second preset difference, the separation state of the steel coil and the coil carrying vehicle is separated; if the first comparison result does not satisfy the first preset condition, and / or the second comparison result does not satisfy the second preset condition, and / or the first difference is less than or equal to the first preset difference, and / or the second difference is less than or equal to the second preset difference, the separation state of the steel coil and the coil carrying vehicle is not separated. The alarm module is configured to perform alarm if the centering state of the steel coil is abnormal.

10. An electronic device, comprising: The electronic device includes: one or more processors; a storage device configured to store one or more programs that, when executed by the one or more processors, cause the electronic device to implement the steel coil centering identification method of any one of claims 1-8.

11. A computer readable storage medium, characterized in that, a computer program stored thereon that, when executed by a processor of a computer, causes the computer to perform the steel coil centering identification method of any one of claims 1-8.

Images