AI visual inspection system, method and equipment for automatically centering coiling height of steel coil and medium

Abstract

The invention provides an AI visual detection system, method and equipment for automatically centering the coiling height of a steel coil and a medium, and the system comprises a visual module which is used for collecting a steel coil image based on a CCD camera; the image processing module is used for carrying out contour detection on the acquired steel coil image; the positioning module is used for identifying edge point coordinates of the steel coil based on contour detection; the offset calculation module is used for calculating the offset of the center position of the steel coil based on the edge point coordinates and the calibration coordinate values of the steel coil; and the calibration module is used for carrying out centering calibration on the steel coil based on the calculated offset of the central position of the steel coil.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of industrial automation control systems, in particular to an AI vision detection system, method, device and medium for automatic centering of the height of a steel coil, and more particularly to application of an intelligent sensing system in an AI vision recognition detection system for positioning of a decoiler on a cold rolling mill. BACKGROUND

[0002] In the process of winding a steel coil, the main shortcomings of the steel coil centering method used in the prior art are as follows: the height measurement system has poor accuracy; the design requires that the maximum deviation of the height centering system from the center line of the coil shaft be ±30mm, but due to equipment aging and other reasons, the design requirement cannot be met. In the production process, the swing arm often collides with the steel coil, increasing the damage to the centering swing arm. According to statistics, there are more than a few major collisions on average every year, each of which affects production for half a day and causes economic losses. Abnormal steel coils cannot be detected. Due to reasons such as excessively wide incoming material and small coil diameter, the steel coil often cannot be wound, and problems such as collision with the coil shaft head occur, causing damage to the steel coil and the coil shaft equipment. Therefore, AI vision detection must be added to ensure the stability and reliability of automatic winding.

[0003] Patent document CN212845076U (application number: 202022032379.8) discloses a workpiece AI vision automatic alignment detection mechanism, which comprises an AI defect detection machine main body, an illumination cylinder lamp and an AI detection head are fixed to the top end of the inner wall of the box of the AI defect detection machine main body, and a through hole is provided on one side of the AI defect detection machine main body, and a supporting plate and a sealing plate are inserted through the through hole, and a workpiece is arranged on the supporting plate, and the supporting plate is provided in an "L" type structure, and the upper end edge of the supporting plate is in close contact with the lower end edge of the workpiece. The workpiece AI vision automatic alignment detection mechanism is provided with a supporting plate, and the edge of the supporting plate can provide good positioning effect for the placement position of the workpiece, so that the detection position of the workpiece is accurate, and the device is provided with a contact sensor, the contact sensor can be triggered by the movable supporting plate to stop the work of the air cylinder, so that the alignment operation of the supporting plate is automated, and good protection is provided for the detection work of the workpiece, and the detection effect is better. SUMMARY

[0004] In view of the defects in the prior art, the purpose of the present application is to provide an AI vision detection system, method, device and medium for automatic centering of the height of a steel coil.

[0005] According to the AI vision detection system for automatic centering of the height of a steel coil provided by the present application, the following is included:

[0006] A vision module is used to collect steel coil images based on a CCD camera;

[0007] An image processing module is used to perform contour detection on the collected steel coil images;

[0008] a positioning module configured to identify edge point coordinates of the steel coil based on the contour detection;

[0009] an offset calculation module configured to calculate a steel coil center position offset based on the edge point coordinates of the steel coil and the calibration coordinate value;

[0010] a calibration module configured to perform steel coil centering calibration based on the calculated steel coil center position offset.

[0011] Preferably, during the process of acquiring the steel coil image based on the CCD camera, a multi-light source combination module is used to uniformly illuminate the surface of the steel coil.

[0012] The multi-light source combination module comprises a ring-shaped LED light source, a polarized light suppression component, and an HDR auxiliary light source.

[0013] The ring-shaped LED light source is configured to provide basic uniform illumination to ensure that the overall brightness of the surface of the steel coil is consistent.

[0014] The polarized light suppression component is configured to eliminate the interference of high light reflection on the metal surface of the steel coil.

[0015] The HDR auxiliary light source is configured to enhance image details by dynamically adjusting exposure.

[0016] Preferably, the offset calculation module comprises: obtaining edge points and a center point of a center axis of a steel coil profile based on the acquired steel coil image; calculating a steel coil center point coordinate based on a plurality of edge points of the steel coil profile; and calculating a difference value between the calculated steel coil center point coordinate and the center point coordinate of the center axis to obtain a steel coil center position offset.

[0017] OffsetX = Xcircle - Xref;

[0018] OffsetY = Ycircle - Yref;

[0019] wherein Xcircle represents the actually detected horizontal coordinate of the center, Ycircle represents the actually detected vertical coordinate of the center, Xref represents the horizontal coordinate of the calibration reference center, Yref represents the vertical coordinate of the calibration reference center, OffsetX represents the horizontal offset value, and OffsetY represents the vertical offset value.

[0020] Preferably, the system further comprises: when the steel coil center position offset exceeds a set threshold value, triggering an alarm and realizing automatic displacement adjustment.

[0021] According to the AI visual detection method for automatic steel coil winding height centering provided by the present application, the method comprises:

[0022] Step S1: the vision module acquires the steel coil image based on the CCD camera;

[0023] Step S2: the image processing module performs contour detection on the acquired steel coil image;

[0024] Step S3: the positioning module identifies the edge point coordinates of the steel coil based on the contour detection;

[0025] Step S4: the offset calculation module calculates the steel coil center position offset based on the edge point coordinates of the steel coil and the calibration coordinate value;

[0026] Step S5: the calibration module performs steel coil centering calibration based on the calculated steel coil center position offset.

[0027] Preferably, in the process of acquiring the steel coil image based on the CCD camera, a multi-light source combination module is used to uniformly illuminate the surface of the steel coil;

[0028] The multi-light source combination module comprises a ring-shaped LED light source, a polarized light suppression component, and an HDR auxiliary light source.

[0029] The ring-shaped LED light source is used to provide basic uniform illumination to ensure consistent overall brightness of the steel coil surface.

[0030] The polarized light suppression component is used to eliminate high light reflection interference on the metal surface of the steel coil.

[0031] The HDR auxiliary light source is used to enhance image details by dynamically adjusting exposure.

[0032] Preferably, the step S4 comprises: the offset calculation module acquires the edge points of the steel coil contour and the center point of the center axis based on the acquired steel coil image; calculates the steel coil center point coordinates based on the multiple edge points of the steel coil contour; calculates the difference value using the calculated steel coil center point coordinates and the center point coordinates of the center axis, and the calculated difference value is the steel coil center position offset.

[0033] OffsetX = Xcircle - Xref;

[0034] OffsetY = Ycircle - Yref;

[0035] Wherein, Xcircle: actual detected center horizontal coordinate; Ycircle: actual detected center vertical coordinate; Xref: horizontal coordinate of the calibration reference center; Yref: vertical coordinate of the calibration reference center; OffsetX: horizontal direction offset value; OffsetY: vertical direction offset value.

[0036] Preferably, the method further comprises: when the steel coil center position offset exceeds a set threshold, triggering an alarm and realizing automatic displacement adjustment.

[0037] According to the electronic device provided by the application, the electronic device comprises a memory and at least one processor, and the memory stores instructions;

[0038] The at least one processor invokes the instructions in the memory to enable the electronic device to perform the steps of the AI visual detection method for automatic centering of the coil height.

[0039] According to the computer readable storage medium provided by the application, the computer readable storage medium stores instructions, and the instructions are executed by a processor to implement the steps of the AI visual detection method for automatic centering of the coil height.

[0040] Compared with the prior art, the application has the following beneficial effects:

[0041] 1. The AI visual detection system, method, device and medium for automatic centering of the coil height cancel the low efficiency caused by personnel-intensive labor, reduce personnel inspection efforts, ensure the safety and stability of the coil during the winding process, and improve the production line running speed; greatly reduce the labor input, and achieve the goal of less people;

[0042] 2. The device can prevent the coil from being turned over and ensure the centering effect of the coil winding, create conditions for unmanned winding of the uncoiler, achieve 100% defect elimination, communicate with the PLC through the TCP / IP communication mode, and realize control interlocking with the L1 system.

[0043] 3. The application captures the edge mark points of the coil profile by a high-resolution camera CCD, finds the characteristic information thereof combined with an image processing algorithm, and calculates the position deviation based on the calibrated results.

[0044] 4. The application can realize accurate centering of the entry step beam area and accurate centering during the winding of the uncoiler, and can solve the problem of centering deviation and coil turning caused by the overflow edge of the incoming hot-rolled strip or the collapse of the inner ring. Since the measurement accuracy is not affected by the overflow edge of the incoming coil head and tail, and can assist daily inspection, the application can be widely applied to cold rolling mill units or other similar working conditions.

[0045] 5. The application can also be applied to other industrial production fields, and only needs to be adjusted according to different processes. BRIEF DESCRIPTION OF DRAWINGS

[0046] Other features, objects and advantages of the application will become more apparent through reading the detailed description of the non-limiting embodiments with reference to the following drawings:

[0047] Fig. 1 Arrangement of workpieces.

[0048] Fig. 2 The schematic diagram for detecting the effect is shown.

[0049] Fig. 3 The control flow chart of the CCD vision system is shown. DETAILED DESCRIPTION

[0050] The application will be described in detail below with specific embodiments. The following embodiments will help those skilled in the art to further understand the application, but do not limit the application in any form. It should be pointed out that, for those skilled in the art, without departing from the concept of the application, a number of changes and improvements can be made. These all belong to the protection scope of the application.

[0051] Example 1

[0052] According to the AI vision detection method for automatically centering the upper coil height of a steel coil provided by the application, as shown in the figure, it comprises: Figs. 1 to 3

[0053] Step S1: The vision module acquires a steel coil image based on a CCD camera;

[0054] Step S2: The image processing module performs contour detection on the acquired steel coil image;

[0055] Step S3: The positioning module identifies the edge point coordinates of the steel coil based on the contour detection;

[0056] Step S4: The offset calculation module calculates the center position offset of the steel coil based on the edge point coordinates of the steel coil and the calibration coordinate value;

[0057] Step S5: The calibration module performs steel coil centering calibration based on the calculated center position offset of the steel coil.

[0058] In this embodiment, in the process of acquiring the steel coil image based on the CCD camera, a multi-light source combination module is used to uniformly illuminate the surface of the steel coil;

[0059] The multi-light source combination module comprises a ring-shaped LED light source, a polarized light suppression component and an HDR auxiliary light source.

[0060] The ring-shaped LED light source is used to provide basic uniform illumination to ensure that the overall brightness of the surface of the steel coil is consistent.

[0061] The polarized light suppression component is used to eliminate the high light reflection interference of the metal surface of the steel coil.

[0062] The HDR auxiliary light source is used to enhance the image details by dynamically adjusting the exposure.

[0063] ​In the embodiment, according to the geometric characteristics, a contour detection technique is used to identify the target object and the mode. This technique has significant effects on image disorder, brightness fluctuation, image blur and object overlap; it can automatically process free-shape objects and has a powerful de-blurring algorithm. For steel coil searching and positioning, it takes no more than 10 ms and can achieve 1 / 40 sub-pixel position repeatability and 0.01 degree rotation repeatability. In addition, it also has rich but easy-to-use image calibration tools. In actual application, through front-mounted CCD vision system, the edge point coordinates of the steel coil are sent to the PLC, and the PLC calculates the deviation value of the steel coil to achieve the purpose of centering calibration.

[0064] The detectable steel coil height range is 1050mm-2050mm.

[0065] The offset calculation module comprises: obtaining the edge points and the center point of the center axis of the steel coil contour based on the collected steel coil image; calculating the steel coil center point coordinates based on the plurality of edge points of the steel coil contour; calculating the difference value by using the calculated steel coil center point coordinates and the center point coordinates of the center axis, and the calculated difference value is the steel coil center position offset, and the steel coil center position offset is converted into a fault signal and sent to L1 and an alarm is given.

[0066] In the embodiment, in combination with the process flow of the cold rolling mill, the specific steps of the contour detection are as follows:

[0067] Before the system is put into production operation, the part most similar to the pre-registered pattern is searched, and the position and inclination thereof are output. The tool is used as a position offset correction source, and the position offset correction of the correction object tool is performed according to the detected changes in the position and inclination of the pattern, and the target positioning and learning are performed according to the correct position of the steel coil in production;

[0068] In automatic production operation, the steel coil is transported to the detection area by the walking beam;

[0069] The steel coil in-place signal is fed back to the L1 control system by the external sensor;

[0070] The vision detection signal is triggered by the L1 control system to obtain the image to be detected;

[0071] The fixed station model is formed by the industrial CCD to capture the surface features of the steel coil. Due to the non-smoothness of the steel coil surface, after being processed by the system controller, unevenness and different peak waveforms will appear, and each waveform value is a characteristic value of the contour detection. The gray value algorithm is used to optimize the curve to obtain a final optimized value (two characteristic values are obtained on both sides of the steel coil).

[0072] The system calculates the interface coordinates of the X axis and the Y axis of the two groups of steel coils respectively according to the optimal characteristic values extracted on both sides of the steel coil.

[0073] In the CCD vision, the whole coordinate point is composed of pixel points, the obtained X\Y coordinate can obtain an actual coordinate value according to the pixel ratio occupied by X\Y, the actual coordinate value is compared with the calibration coordinate, so that the X\Y deviation value of the steel coil is calculated, and the accurate positioning of the steel coil is realized.

[0074] The application further provides an AI vision detection system for automatic centering of steel coil winding height, which can be realized by executing the process steps of the AI vision detection method for automatic centering of steel coil winding height, that is, the AI vision detection method for automatic centering of steel coil winding height can be understood as the preferred implementation of the AI vision detection system for automatic centering of steel coil winding height by those skilled in the art.

[0075] The embodiment uses the PLC of SIEMENS, a series of detections are completed by automatically issuing commands through the L1 system to realize full automation and high detection rate of the equipment, and the current steel coil running state is fed back through WINCC alarm indication.

[0076] Those skilled in the art know that, in addition to implementing the system, device and each module thereof provided by the application in the form of pure computer readable program code, the same program can also be realized in the form of logic gate, switch, application specific integrated circuit, programmable logic controller and embedded microcontroller by logically programming the method steps. Therefore, the system, device and each module thereof provided by the application can be considered as a hardware component, and the modules included therein for realizing various programs can also be considered as structures in the hardware component; the modules for realizing various functions can also be considered as both software programs for realizing methods and structures in the hardware component.

[0077] The specific embodiments of the application are described above. It should be understood that the application is not limited to the above specific embodiments, and those skilled in the art can make various changes or modifications within the scope of the claims, which does not affect the essential content of the application. In the case of no conflict, the embodiments of the application and the features in the embodiments can be combined with each other arbitrarily.

Claims

1. An AI visual inspection system for automatic centering of coil height, characterized by: include: Vision module, used to capture steel coil images based on CCD camera; Image processing module, used to perform contour detection on the collected steel coil images; Positioning module, used to identify the edge coordinates of the steel coil based on contour detection; An offset calculation module is used to calculate the offset of the center position of the steel coil based on the edge point coordinates of the steel coil and the calibration coordinate values; The calibration module is used to calibrate the center of the steel coil based on the calculated offset of the center position of the steel coil.

2. The AI ​​visual inspection system for automatic centering of coil height according to claim 1 is characterized in that: In the process of collecting the steel coil image based on the CCD camera, a multi-light source combination module is used to uniformly illuminate the surface of the steel coil; The multi-light source combination module includes: a ring-shaped LED light source, a polarized light suppression component and an HDR auxiliary light source; The annular LED light source is used to provide basic uniform lighting to ensure uniform brightness on the entire surface of the steel coil; The polarized light suppression component is used to eliminate the interference of high light reflection on the metal surface of the steel coil; The HDR auxiliary light source is used to enhance image details by dynamically adjusting exposure.

3. The AI ​​visual inspection system for automatic centering of coil height according to claim 1 is characterized in that: The offset calculation module includes: obtaining edge points of the steel coil contour and the center point of the central axis based on the collected steel coil image; calculating the coordinates of the center point of the steel coil based on the multiple edge points of the steel coil contour; and calculating the difference between the calculated coordinates of the center point of the steel coil and the coordinates of the center point of the central axis, and the calculated difference is the offset of the center position of the steel coil; OffsetX = Xcircle - Xref; OffsetY=Ycircle-Yref; Among them, Xcircle: the horizontal coordinate of the circle center actually detected; Ycircle: the vertical coordinate of the circle center actually detected; Xref: the horizontal coordinate of the calibration reference center; Yref: the vertical coordinate of the calibration reference center; OffsetX: the horizontal offset value; OffsetY: the vertical offset value.

4. The AI ​​visual inspection system for automatic centering of coil height according to claim 1 is characterized in that: The system also includes: when the offset of the center position of the steel coil exceeds a set threshold, an alarm is triggered and automatic displacement adjustment is achieved.

5. An AI visual inspection method for automatic centering of coil height, characterized in that: include: Step S1: The vision module collects the steel coil image based on the CCD camera; Step S2: The image processing module performs contour detection on the collected steel coil image; Step S3: The positioning module identifies the edge point coordinates of the steel coil based on contour detection; Step S4: The offset calculation module calculates the offset of the center position of the steel coil based on the edge point coordinates of the steel coil and the calibration coordinate values; Step S5: The calibration module performs steel coil centering calibration based on the calculated steel coil center position offset.

6. The AI ​​visual inspection method for automatic centering of coil height according to claim 5 is characterized in that: In the process of collecting the steel coil image based on the CCD camera, a multi-light source combination module is used to uniformly illuminate the surface of the steel coil; The multi-light source combination module includes: a ring-shaped LED light source, a polarized light suppression component and an HDR auxiliary light source; The annular LED light source is used to provide basic uniform lighting to ensure uniform brightness on the entire surface of the steel coil; The polarized light suppression component is used to eliminate the interference of high light reflection on the metal surface of the steel coil; The HDR auxiliary light source is used to enhance image details by dynamically adjusting exposure.

7. The AI ​​visual inspection system for automatic centering of coil height according to claim 5 is characterized in that: Step S4 includes: an offset calculation module acquiring edge points of the steel coil contour and a center point of the central axis based on the collected steel coil image; calculating the coordinates of the center point of the steel coil based on the multiple edge points of the steel coil contour; and calculating a difference between the calculated coordinates of the center point of the steel coil and the coordinates of the center point of the central axis, where the calculated difference is the offset of the center position of the steel coil; OffsetX = Xcircle - Xref; OffsetY=Ycircle-Yref; Among them, Xcircle: the horizontal coordinate of the circle center actually detected; Ycircle: the vertical coordinate of the circle center actually detected; Xref: the horizontal coordinate of the calibration reference center; Yref: the vertical coordinate of the calibration reference center; OffsetX: the horizontal offset value; OffsetY: the vertical offset value.

8. The AI ​​visual inspection method for automatic centering of coil height according to claim 5 is characterized in that: The method further includes: when the offset of the center position of the steel coil exceeds a set threshold, an alarm is triggered and automatic displacement adjustment is achieved.

9. An electronic device comprising a memory and at least one processor, wherein the memory stores instructions; The at least one processor calls the instructions in the memory to enable the electronic device to execute each step of the AI ​​visual detection method for automatic centering of the coil height of the steel coil as described in any one of claims 1 to 4.

10. A computer-readable storage medium having instructions stored thereon, characterized in that: When the instructions are executed by the processor, the various steps of the AI ​​visual detection method for automatic centering of the coil height as described in any one of claims 1 to 4 are implemented.

Images