Panel material edge defect detection system
By designing the edge defect detection system of panel material and using imaging and sensor technology to achieve automated detection, the problem of difficult edge defects in the prior art is solved, and detection accuracy and production efficiency are improved.
Patent Information
- Application Number
- CN201911177658.9
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2019-11-27
- Publication Date
- 2024-11-29
- Estimated Expiration
- 2039-11-27
AI Technical Summary
The prior art is difficult to automatically detect edge defects of panel materials, resulting in high labor intensity and high missed detection rate, which affects product quality and production efficiency.
Design a panel material edge defect detection system, including a detection device, lighting device, transmission device, controller and digital image processor, identify edge defects through imaging scanning and image analysis, and obtain multi-angle using lens imaging and mirror imaging devices Images, combined with distance measurement and speed measurement sensors to achieve dynamic detection.
It improves detection accuracy and efficiency, reduces the labor intensity of manual inspection, can timely identify edge defects of board materials, reduces product recall and rework rates, and optimizes production processes.
Smart Images

Figure CN110793974B_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of plate material edge defect detection, and in particular to a plate material edge defect detection system. Background Art
[0002] During the processing of various panel materials, such as float glass, core board, steel plate, PVC board, etc., only the upper and lower surfaces are generally inspected for defects, and it is difficult to identify whether there are defects on the edges. Manual identification has the defects of high labor intensity and high missed detection rate. In addition, there are often quality problems such as scratches and wrinkles on the edges of the panel materials during the production and processing process, which affects the subsequent application of the product. If the defects are not discovered in time, it is easy to lead to product recall or high repair rate, which is not conducive to the timely optimization of the production line. Therefore, it is urgent to design a system that can detect edge defects of panel materials. Summary of the invention
[0003] The invention provides a plate material edge defect detection system to solve the technical problem that the existing plate material cannot automatically detect its edge defects.
[0004] According to one aspect of the present invention, a plate material edge defect detection system is provided, comprising at least one set of detection devices, arranged on the outer peripheral side of the plate material, for imaging and scanning the edge of the plate material,
[0005] At least one lighting device for providing lighting for imaging of the panel material,
[0006] The conveying device is used to make the board material and the detection device move relative to each other.
[0007] A controller is communicatively connected to the detection device and the lighting device, and is used to control the switching of the lighting mode and control the detection device to continuously scan the edge of the board material to obtain image data under the corresponding lighting mode.
[0008] a digital image processor, communicatively connected to the controller and the detection device, for generating control instructions to the controller and receiving image data generated by the detection device and performing edge defect analysis based on the received image data,
[0009] The detection device includes a lens imaging device, a mirror imaging device and a machine frame. The machine frame is a one-way opening structure. The opening end of the machine frame is arranged relative to the edge of the board material. The mirror imaging device and the lens imaging device are arranged relatively to each other in the machine frame. The mirror imaging device is used to form a virtual image of the upper edge and the lower edge of the board material, and the lens imaging device is used to obtain a real image of the edge of the board material and a virtual image formed in the mirror imaging device.
[0010] Further, the mirror imaging device is located at the open end of the frame and is respectively arranged on the upper and lower sides of the frame and forms a certain angle, and the lens imaging device is located at the closed end of the frame and is arranged at the bisector of the angle; or
[0011] The mirror imaging device is located at the closed end of the frame and is arranged at a certain angle, and the lens imaging device is located at the open end of the frame and is located above or below the frame and is located on the angle bisector of the mirror imaging device.
[0012] Furthermore, it also includes
[0013] Distance sensor, used to measure the distance between the edge of the board material and the reference object.
[0014] Speed sensor, used to measure the moving speed of the board material.
[0015] The pan-tilt device is used to maintain the distance between the detection device and the edge of the panel material.
[0016] The distance measuring sensor and the speed measuring sensor are communicatively connected to the digital image processor, and the pan-tilt device is communicatively connected to the controller. The digital image processor generates a control instruction according to the detection results of the distance measuring sensor and the speed measuring sensor and transmits the control instruction to the controller. The controller controls the pan-tilt device to dynamically maintain the distance between the detection device and the edge of the board material according to the received control instruction.
[0017] Furthermore, the pan-tilt device is also used to adjust the angle of the detection device relative to the edge of the board material. The digital image processor generates an angle adjustment control instruction for the detection device based on the received image clarity and transmits it to the controller. The controller controls the pan-tilt device to adjust the angle of the detection device relative to the edge of the board material based on the received angle adjustment control instruction.
[0018] Furthermore, the lens imaging device and the mirror imaging device are both provided with a gimbal and are communicatively connected with a controller, and the controller controls the gimbal to adjust the angle of the lens imaging device and / or the mirror imaging device.
[0019] Furthermore, there are multiple sets of the detection devices, which are respectively arranged corresponding to multiple edges of the board material.
[0020] Furthermore, the lighting device is arranged in the frame.
[0021] Furthermore, the lighting device is a combined light source including a plurality of light sources, and the combined light source provides a plurality of lighting modes with different lighting brightness by time-sharing switching lighting between different combinations of the light sources under the control of the controller.
[0022] Furthermore, the digital image processor comprises:
[0023] The defect recognition module is used to obtain the image data of each channel through a segmentation algorithm and reorganize it according to the imaging angle to obtain independent images of each angle, and perform pattern matching with the image of the standard panel material to identify defects.
[0024] Furthermore, the digital image processor further includes at least one of the following:
[0025] The quality grade classification module is used to collect statistics on defect analysis results to classify product quality and obtain quality grade information;
[0026] A defect position analysis module is used to determine the length, width, area and three-dimensional relative position information of each defect according to the defect analysis results, and generate a plane distribution map and / or a three-dimensional distribution map of the plate type defects;
[0027] The storage module is used to receive and store the defect analysis results.
[0028] The present invention has the following beneficial effects:
[0029] The panel material edge defect detection system of the present invention uses a lens imaging device to capture the real image of the panel material edge surface under different lighting modes and the virtual image formed by the mirror imaging device, thereby achieving multi-angle images of the panel material edge at a fixed imaging angle of view, and detecting and identifying edge defects of the panel material. The detection device integrates a lens imaging device and at least two mirror imaging devices in a machine frame. The structure of the detection device is simple, and the integration of the detection device is improved. The minimum lens imaging device and mirror imaging device are used to achieve multi-angle images at a fixed imaging angle of view, and the edge defects of the panel material are detected panoramically. At the same time, the imaging angle of view, logical position, time, imaging environment and other factors of the detection device are unified, which overcomes the complexity and uncertainty caused by external environmental interference, focusing, time changes, etc., and the detection accuracy and ease are higher.
[0030] In addition, for the lens imaging device deployed on the bisector of the angle formed by the mirror imaging device, the distance to the real image at the edge of the board material and the virtual image formed in the mirror imaging device is basically equal. It does not need to focus, and clear images of the real images at different angles at the same position of the edge of the board material and the virtual image formed in the mirror imaging device can be obtained at the same time, effectively avoiding the difference in imaging time at the same position due to focusing and the complexity and uncertainty caused by the difference in angle.
[0031] In addition, by setting a distance measuring sensor to detect the distance between the edge of the board material and the reference object in real time and using a speed measuring sensor to detect the moving speed of the board material in real time, the digital image processor generates a control instruction based on the detection results of the distance measuring sensor and the speed measuring sensor and transmits it to the controller. The controller controls the pan-tilt device in real time to dynamically maintain the distance between the detection device and the edge of the board material according to the received control instruction, thereby realizing dynamic detection and improving the accuracy under dynamic detection.
[0032] In addition, the digital image processor can also generate angle adjustment control instructions for the detection device based on the acquired image clarity. The controller controls the pan-tilt device to adjust the angle of the detection device relative to the edge of the board material accordingly according to the angle adjustment control instructions, so that the lens imaging device can always obtain clear real and virtual images of the edge of the board material, further improving the accuracy of detection.
[0033] In addition to the above-described purposes, features and advantages, the present invention has other purposes, features and advantages. The present invention will be further described in detail with reference to the accompanying drawings. BRIEF DESCRIPTION OF THE DRAWINGS
[0034] The drawings constituting a part of this application are used to provide a further understanding of the present invention. The exemplary embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation of the present invention. In the drawings:
[0035] Figure 1 It is a structural schematic diagram of a panel material edge defect detection system according to a preferred embodiment of the present invention.
[0036] Figure 2 It is a preferred embodiment of the present invention Figure 1 Schematic diagram of the structure of the detection device.
[0037] Description of Reference Numerals
[0038] 1. Distance sensor; 2. Speed sensor; 3. Combined light source; 4. Bracket; 5. Detection device; 6. Detected panel material; 7. Transmission device; 8. PTZ device; 9. Controller; 10. Digital image processing device; 11. Data communication cable; 12. Control communication cable; 13. Camera; 14. Low-transmittance coated reflector; 15. Machine frame. DETAILED DESCRIPTION
[0039] The embodiments of the present invention are described in detail below with reference to the accompanying drawings. However, the present invention can be implemented in many different ways as defined and covered below.
[0040] It should be understood that the drawings and descriptions of the present invention have been simplified to help clearly understand the system and various components of the present invention by example, and non-critical components in the panel material edge defect detection system have been removed for the purpose of focusing on the description of the present invention. Those skilled in the art will recognize that other components may be desirable and / or necessary to implement the present invention. However, since these components are well known in the art and since they do not help to better understand the present invention, descriptions of these components are not given herein. It should also be understood that the drawings included herein only provide graphical representations of current examples of the present invention, and structures falling within the scope of the present invention may include structures different from those shown in these drawings. In the drawings, similar structures are given similar figure marks.
[0041] This embodiment provides a panel material edge defect detection system for detecting edge defects of panel materials. The panel material can be glass, steel plate, tile, wood board or other panel materials, or other profiles with geometric panels, such as float glass strip, mobile phone panel glass, steel plate, aluminum alloy profile, etc. In the example, it is agreed to be a rectangular panel material. The panel material edge defect detection system includes:
[0042] At least one set of detection devices, arranged on the outer peripheral side of the board material, for imaging and scanning the edge of the board material,
[0043] At least one lighting device for providing lighting for imaging of the panel material,
[0044] a conveying device for moving the panel material relative to the detection device,
[0045] A controller is communicatively connected to the detection device and the lighting device, and is used to control the switching of the lighting mode and control the detection device to continuously scan the edge of the board material to obtain image data under the corresponding lighting mode.
[0046] a digital image processor, communicatively connected to the controller and the detection device, for generating control instructions to the controller and receiving image data generated by the detection device and performing edge defect analysis based on the received image data,
[0047] The detection device includes a lens imaging device, a mirror imaging device and a machine frame. The machine frame is a one-way opening structure. The opening end of the machine frame is arranged relative to the edge of the board material. The mirror imaging device and the lens imaging device are arranged relatively to each other in the machine frame. The mirror imaging device is used to form a virtual image of the upper edge and the lower edge of the board material, and the lens imaging device is used to obtain a real image of the edge of the board material and a virtual image formed in the mirror imaging device.
[0048] In this embodiment, there can be multiple sets of detection devices, and the multiple sets of detection devices are respectively set corresponding to multiple edges of the plate material. By setting detection devices at different edge positions of the plate material, multiple edges of the plate material can be detected at the same time, which is better adapted to edge defect detection of plate materials with complex geometric shapes.
[0049] In this embodiment, the lighting device is composed of a plurality of light sources, which are arranged in a display or circular manner. The combined light source can provide a plurality of lighting modes with different lighting brightness by switching the lighting in different combinations between the light sources under the control of the controller. Preferably, the lighting device can also perform a plurality of spectrum conversions under the control of the controller. If the optional number of spectrum conversions is S, the total lighting mode of the lighting device will increase by S times to meet the requirements of different tested panel materials for lighting brightness and spectrum. It can be understood that the combined light source can preferably be an LED light strip, which has a long life and can obtain higher brightness.
[0050] In this embodiment, the conveying device can drive the plate material to move relative to the detection device or drive the detection device to move relative to the plate material. In order to facilitate control and simplify the structure, it is preferred that the conveying device drives the plate material to move relative to the detection device. It can be understood that the plate material edge defect detection system also includes a supporting device, which is used to install and fix the detection device. The plate material is set on the conveying device, so that the detection device does not move, and the conveying device drives the plate material to move relative to the detection device. In addition, the lighting device can also be set on the supporting device.
[0051] In this embodiment, the controller can control the switching of lighting modes, specifically by controlling the combined light source to switch the lighting in time according to the required combination. The controller can also control the detection device to continuously scan the edge of the board material when it is illuminated to obtain imaging data of each channel.
[0052] In this embodiment, the digital image processor can generate a control instruction for combining the illumination modes of the light source and transmit it to the controller, and the controller controls the switching of the illumination modes according to the received control instruction, so that the illumination matrix switches the illumination in a required combination in a time-sharing manner. The digital image processor can also receive image data generated by at least one detection device, and reorganize the image data of the imaging device in a row of the imaging matrix according to the illumination mode switched in a time-sharing manner, so that the imaging matrix forms an independent image for each illumination mode, and performs defect recognition, statistical analysis, etc. on the defect information of the obtained image.
[0053] In this embodiment, the detection device integrates the lens imaging device and the mirror imaging device in a frame. The shape of the detection device can be a cylinder, a cone, a rotating body, a cross-section body, etc. The frame is a one-way opening structure. The opening end of the frame is arranged relative to the edge of the board material. The mirror imaging device and the lens imaging device are arranged relatively in the frame. The mirror imaging device can form a virtual image of the upper edge and the lower edge of the board material. The lens imaging device can obtain a clear real image of the edge of the board material and a virtual image formed in the mirror imaging device. The mirror imaging device can be a low-transmittance coated reflector, or a prism with one or several sides coated with a low-transmittance film. Specifically, the number of the mirror imaging devices is at least two, the mirror imaging device is located at the open end of the frame, and is respectively arranged on the upper and lower sides of the frame and arranged at a certain angle, the lens imaging device is located at the closed end of the frame and is arranged at the bisector of the angle, or the mirror imaging device is located at the closed end of the frame and is arranged at a certain angle, the lens imaging device is located at the open end of the frame and is located above or below the frame, and is located near the bisector of the angle of the mirror imaging device. The lens imaging device deployed here has a substantially equal distance to the real image at the edge of the board material and the virtual image formed in the mirror imaging device. It does not need to adjust the focal length, and can simultaneously obtain clear images of the real image at different angles at the same position of the board material edge and the virtual image formed in the mirror imaging device, effectively avoiding the complexity and uncertainty caused by the difference in imaging time at the same position due to focusing and the difference in angle. The detection device integrates a lens imaging device and at least two mirror imaging devices in a frame, and through the setting of special positions, it can obtain clear real and virtual images of an edge of the board material, so that defect identification and analysis can be performed on the edge.
[0054] The board material edge defect detection system of this embodiment only uses a lens imaging device to capture the real image of the board material edge surface under different lighting modes and the virtual image formed by the mirror imaging device, thereby achieving multi-angle images of the board material edge at a fixed imaging angle of view, and detecting and identifying edge defects of the board material. In addition, by integrating a lens imaging device and at least two mirror imaging devices in a machine frame, the structure of the detection device is simple, and the integration of the detection device is improved. With a minimum of lens imaging devices and mirror imaging devices, multi-angle images can be obtained at a fixed imaging angle of view, and edge defects of the board material can be detected panoramically. At the same time, due to the unification of factors such as imaging angle of view, logical position, time, and imaging environment, the complexity and uncertainty brought by the image reconstruction of multi-angle imaging are overcome, and the investment cost is lower.
[0055] It can be understood that, as a preferred embodiment, the panel material edge defect detection system further includes:
[0056] Distance sensor, used to measure the distance between the edge of the board material and the reference object.
[0057] Speed sensor, used to measure the moving speed of the board material.
[0058] A pan / tilt device, used to adjust the angle of the detection device relative to the edge of the panel material and / or to maintain the distance between the detection device and the edge of the panel material,
[0059] The distance measuring sensor and the speed measuring sensor are communicatively connected to the digital image processor, and the pan-tilt device is communicatively connected to the controller. The digital image processor generates a control instruction according to the detection results of the distance measuring sensor and the speed measuring sensor and transmits the control instruction to the controller. The controller controls the pan-tilt device to dynamically maintain the distance between the detection device and the edge of the board material according to the received control instruction.
[0060] Specifically, the distance measuring sensor, the speed measuring sensor and the pan-tilt device are all installed on the supporting device, and the detection device is installed on the pan-tilt device.
[0061] In this preferred embodiment, the digital image processor obtains the distance between the edge of the board material and the reference object in real time through the distance measuring sensor and obtains the moving speed of the board material in real time through the speed measuring sensor, and performs analysis and calculation to obtain the dynamic distance and adjustment speed between the detection device and the board material, and generates a control instruction to the controller. The controller controls the pan-tilt device in real time to dynamically maintain the distance between the detection device and the edge of the board material according to the received control instruction, thereby realizing dynamic detection. In addition, the digital image processor can also analyze the clarity of the acquired image, and generate an angle adjustment control instruction for the detection device to the controller. The controller controls the pan-tilt device to correspondingly adjust the angle of the detection device relative to the edge of the board material according to the received control instruction, so that the lens imaging device can obtain clear real and virtual images.
[0062] The board material edge defect detection system of this preferred embodiment detects the distance between the edge of the board material and the reference object in real time by setting a distance sensor and detecting the moving speed of the board material in real time by using a speed sensor. The digital image processor generates a control instruction based on the detection results of the distance sensor and the speed sensor and transmits it to the controller. The controller controls the pan-tilt device in real time to dynamically maintain the distance between the detection device and the edge of the board material according to the received control instruction, thereby realizing dynamic detection and improving the accuracy under dynamic detection. In addition, the digital image processor can also generate an angle adjustment control instruction for the detection device based on the acquired image clarity. The controller controls the pan-tilt device to correspondingly adjust the angle of the detection device relative to the edge of the board material according to the angle adjustment control instruction, so that the lens imaging device can always obtain a clear real image and virtual image of the edge of the board material, further improving the detection accuracy.
[0063] It can be understood that, as a further preferred embodiment, the lens imaging device and the mirror imaging device are both equipped with a gimbal, and the gimbals of the lens imaging device and the mirror imaging device are both communicatively connected to the controller. The digital image processor can obtain angle adjustment control instructions for the lens imaging device and / or the mirror imaging device based on the received image clarity analysis and send them to the controller. The controller controls the gimbal of the lens imaging device and / or the mirror imaging device to adjust the angle of the lens imaging device and / or the mirror imaging device according to the received control instructions, thereby realizing separate angle adjustment or simultaneous adjustment of the lens imaging device and the mirror imaging device, facilitating refined control and further improving the clarity of imaging and the accuracy of detection.
[0064] It can be understood that, as a preferred embodiment, the lighting device can be set in the frame of the detection device, thereby further improving the integration of the detection device and reducing the difficulty of engineering implementation. Alternatively, the lighting device is set in the frame and on the support device at the same time, thereby achieving multi-angle lighting, providing better lighting effects, and further improving imaging clarity.
[0065] It can be understood that the digital image processor includes a defect recognition module, which is used to obtain independent images of each angle by segmenting the image data of each channel and reorganizing them according to the imaging angle, and perform pattern matching with the image of the standard board material to identify defects.
[0066] The defect recognition module divides each frame of the image through a segmentation algorithm, reorganizes the independent images at each angle according to the imaging angle, and then performs image analysis. It identifies edge defects through defect analysis methods such as pattern matching with the image of the edge of a defect-free standard board material, and calculates the length, width and area of each defect. The three-dimensional position of each defect in the edge of the board material is calculated based on the number of pixels at the boundary between the defect and the edge of the board material.
[0067] Preferably, the digital image processor also includes a storage module for receiving and storing the analysis results of the defect recognition module, so as to facilitate the control system of the subsequent production line to optimize and transform the process based on the analysis results.
[0068] Preferably, the digital image processor also includes a quality grade classification module, which is used to perform statistics on the defect analysis results to perform quality grade classification on the products to obtain quality grade information; statistical analysis is performed based on the detection results to accurately classify the defect categories of the board materials, such as pits, convex points, scratches, cracks, etc., and the products are graded in turn, the density of defects at each level is calculated, and the quality grade information is sent in coordination with the corresponding coding equipment, and the coding is performed on the edge surface of the board material. The online robot in the subsequent link directly reads the code, performs intelligent cutting, removes the deformed segments, and performs intelligent sorting.
[0069] Preferably, the digital image processor further comprises a defect position analysis module, which is used to determine the length, width, area and three-dimensional relative position information of each defect according to the defect analysis results, and generate a plane distribution map and / or a three-dimensional distribution map of the plate shape defects; for example, based on the acquired defect data information, defect length, width, area and three-dimensional relative position information, directly draw a plane distribution map of the plate shape defects. Preferably, if a three-dimensional modeling graphic of a transparent material is imported, a three-dimensional distribution map of the plate shape defects is drawn.
[0070] Figure 1 The preferred embodiment of the detection system for detecting edge defects of the panel material 6 to be tested is shown. It includes:
[0071] Distance sensor 1;
[0072] Speed sensor 2;
[0073] Lighting device: combined light source 3;
[0074] Support device: bracket 4;
[0075] The detection device 5 includes a camera 13 (lens imaging device), a low-transmittance coated reflector 14 (mirror imaging device), and a frame 15;
[0076] Tested panel material6.
[0077] Conveying device 7;
[0078] PTZ device 8;
[0079] Controller 9;
[0080] Digital image processor 10.
[0081] In the present invention, the panel material 6 to be tested can be glass, steel plate, tile, wood board or other panel materials, or other panel profiles containing geometric shapes, such as float glass strip, mobile phone panel glass, steel plate, aluminum alloy profile, etc. In the example, it is agreed to be a rectangular panel material.
[0082] The conveying device 7 is used to allow the measured panel material 6 and the bracket 4 to move relative to each other. Figure 1As shown in , the above relative movement is generated by moving the panel material 6 to be tested relative to the bracket 4. The above relative movement can also be obtained by moving the bracket 4 relative to the panel material 6 to be tested. For example, when the panel material 6 to be tested has a large mass, a large size, and a special shape, it is easier to move the bracket 4 than to move the panel material 6 to be tested. The conveying device 7 in the present invention includes, for example, a slide, a stepper motor, a conveyor belt, a support frame, etc. For the purpose of illustration and not limitation, it will be assumed hereinafter that the bracket 4, the detection device 5, the distance sensor 1, the speed sensor 2, and the pan-tilt device 8 remain fixed, and the panel material 6 to be tested moves relative to them.
[0083] Figure 1 The distance sensor 1, speed sensor 2, combined light source 3, bracket 4, detection device 5, conveying device 7, pan-tilt device 8, controller 9, digital image processor 10 in the board material edge defect detection system are shown, as well as their relative positional relationship with the board material 6 to be tested. The controller 9 is connected to the combined light source 3, the detection device 5, and the pan-tilt device 8 through the control communication cable 12, and the digital image processor 10 is connected to the controller 9, the distance sensor 1, the speed sensor 2, the detection device 5, and the pan-tilt device 8 through the data communication cable 11. In the board material edge defect detection system, the board material 6 to be tested moves at a constant speed V along the vertical direction of the plane. The detection device 5 in this example is installed on the opposite side, and one is deployed in each example. In actual application, the number can be reasonably configured according to the detection needs. If two more detection devices 5 are added, they can be deployed on two mutually perpendicular conveyor belts to realize the full edge defect detection of rectangular board materials, that is, the front, back, left and right edges of the board material are detected simultaneously. The camera 13 is responsible for collecting light, imaging the collected light onto its photosensitive surface, and converting it into an electrical signal. In this example, the lens imaging device can use a CCD linear array imaging component, a CMOS linear array imaging component, or other linear array imaging components, and can usually integrate corresponding image processing functions and output relevant defect data information. According to the detection needs of the board material 6 to be tested, the camera 13 can be installed above or below the frame 15, and can also be installed behind the frame 15. Figure 2 The situations where the camera 13 is located above and behind the machine frame 15 are respectively shown. For the transparent panel material 6 to be tested, when the camera 13 is installed above or below the machine frame 15, the real image of the edge side, the lower corner area of the edge or the upper corner area of the edge can be captured from the rear of the edge of the panel material 6 to be tested, and the virtual image of the lower corner area and the upper corner area of the edge can also be captured from the low-transmittance coated reflector 14.
[0084] like Figure 1As shown, the distance sensor 1 transmits the distance L collected from the edge of the tested board material 6 and the reference object to the digital image processor 10 in real time, and the speed sensor 2 transmits the transmission speed V collected from the tested board material 6 to the digital image processor 10 in real time. The digital image processor 10 analyzes and calculates the distance L and speed V, determines the dynamic distance and adjustment speed between the detection device 5 and the edge of the tested board material 6, and forms a control instruction to the controller 9. The controller 9 dynamically controls the pan-tilt device 8 in real time to dynamically maintain the distance between the detection device 5 and the edge of the tested board material 6, so as to dynamically adapt to the dynamic detection of defects of the tested board material 6 under the condition of dynamic changes in the distance between the edge of the tested board material 6 and the reference object of various geometric shapes. The digital image processor 10 can analyze the clarity of the acquired image, generate the angle information of the camera 13 and the low-transmittance coated reflector 14 to the controller 9, and the controller 9 controls the pan-tilt device 8 to adjust the angle of the camera 13 and the low-transmittance coated reflector 14, so as to obtain clear real and virtual images with the camera 13. The digital image processor 10 can also generate control instructions for the lighting mode of the combined light source 3. The controller 9 controls the switching of the lighting mode of the lighting device 3 according to the control instructions, so that the lighting matrix switches the lighting in time according to the required combination, and controls the detection device 5 to continuously scan the edge of the board material 6 to be tested when it is illuminated, and obtains imaging data of each channel.
[0085] like Figure 1 As shown, in the example of the present invention, the combined light source 3 is composed of a plurality of light sources and is arranged above the board material 6 to be tested. When the combined light source 3 is turned on, the light is emitted and irradiated onto the board material 6 to be tested, and enters the camera 13 after being reflected and / or transmitted by the edge of the board material. Due to the bright lighting environment, the distortion and refraction unevenness in the edge of the board material 6 to be tested are very sensitive, forming an image completely different from the non-defective area, so that defects such as concave points, convex points, and deformation of the board material 6 to be tested can be detected accordingly. At the same time, under the control of the controller 9, the combined light source 3 is turned on or off according to the required combination to provide a lighting environment with a variety of lighting brightness and / or spectrum. Unless otherwise specified, in this embodiment, when a certain light source or several light sources in a certain combined light source are turned on, the other light sources are turned off.
[0086] Continue to refer to Figure 1 As shown, the combined light source 3 is turned on, and after the light is emitted, it is irradiated onto the board material 6 to be tested. The camera 13 can directly collect the image on the edge of the board material 6 to be tested. If the board material 6 to be tested is transparent, the real image of the edge side, the lower corner area of the edge or the upper corner area of the edge can be captured from behind the edge. At the same time, the camera 13 can also capture the virtual image of the lower corner area and the upper corner area of the edge of the board material 6 to be tested from the low-transmittance coated reflector 14.
[0087] Experiments show that the digital image processor 10 divides the obtained image, reorganizes the independent images at each angle according to the imaging angle, and then performs image analysis and pattern matching with the image of the edge of the non-standard board material to identify defects and calculate the length, width and area of each defect. The three-dimensional position of each defect in the edge of the board material is calculated by the number of pixels at the boundary between the defect and the edge of the board material. The digital image processor 10 performs statistical analysis on the defect data, accurately classifies the defect types at the edge of the board material, such as pits, convex spots, scratches, cracks, etc., performs quality grade on the products in turn, calculates the density of defects at each level, and cooperates with the corresponding coding equipment to send quality grade information and code on the surface of the board material. The online robot in the subsequent link directly reads the code, performs intelligent cutting, removes the deformed segments, and intelligently sorts; the analysis results are used to intelligently evaluate the quality stability of the production line at the edge of the board material, and accordingly optimize and transform the relevant production processes; the obtained defect data information, defect length, width, area and three-dimensional relative position information are used to directly draw a planar distribution map of the board shape defects. If the three-dimensional modeling graphics of the edge of the board material are imported, a three-dimensional distribution map of the board shape defects can also be drawn.
[0088] It is particularly noted that the light source in the present invention can be a semiconductor light source or an ordinary light source; the spectral range is not limited, but it must be within the photosensitivity range of the imaging device; the light source can be monochromatic light or white light. In this example, the light sources of the combined light source 3 are not turned on at the same time, but the combined light source is switched by the controller 9 in time-sharing mode to illuminate the board material 6 under test, and the camera 13 continuously scans in real time, alternately acquiring defect data information under each lighting mode. Of course, the alternation of light source modes does not exhaust all light source combinations, nor does it need to be in a fixed order, but rather selects several optimal lighting modes according to the detection requirements, and switches the lighting cyclically. In order to control the working sequence of the camera 13 and the low-transmittance coated reflector 14, in Figure 1 A controller 9 is provided in the board material edge defect detection system. The controller 9 is used as an external trigger source to control the trigger timing of each of the lighting device and the lens imaging device. The camera 13 and the controller 9 may include any type of pulse trigger, such as but not limited to an encoder. During the detection process, the camera 13 and the controller 9 sense the displacement of the board material 6 to be tested and control the operation of each light source and the lens imaging device, so that all channels are detected once within one working cycle.
[0089] The board material edge defect detection system of the preferred embodiment adopts dual detection devices to achieve panoramic image acquisition of board material edge with a fixed imaging angle of view with a minimum of lens imaging devices, and all-round and simultaneous detection of various board material edge defects. The uniformity of imaging angle of view, logical position, time, imaging environment and other factors overcomes the complexity and uncertainty brought by multiple imaging angles and image reconstruction. Compared with multi-angle views and multi-imaging matrices, the system structure is simpler, with fewer lens imaging devices and lower costs.
[0090] It should be noted that the above examples of the present invention are only for the purpose of illustration and explanation, and are not intended to limit the present invention to the specific form disclosed. Those skilled in the art can fully conceive of various forms of modifications and variations by reading this specification. For example, in the edge defect detection system of the plate material of the present invention, the detection device is not limited to one, but can be multiple. The increase of the detection device, assisted by the change of the deployment orientation, can realize the full edge detection of the plate material, and at the same time, it will also multiply the scanning frequency and the amount of defect information. The defect information of each detection device complements each other, corroborates each other, and checks each other, so that the defect shape, size, position and other indicator information are more accurate. Compared with a single detection device, the defect recognition rate and detection rate are higher. Due to the change in the orientation of the detection device, it is better adapted to the detection of profiles with complex geometric shapes, such as aluminum alloy profiles.
[0091] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. For those skilled in the art, the present invention may have various modifications and variations. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.
Claims
1. A panel material edge defect detection system, characterized in that: It includes at least one set of detection devices, which are arranged on the outer peripheral side of the board material and are used to perform imaging scanning on the edge of the board material. At least one lighting device for providing lighting for imaging of the panel material, The conveying device is used to make the board material and the detection device move relative to each other. A controller is communicatively connected to the detection device and the lighting device, and is used to control the switching of the lighting mode and control the detection device to continuously scan the edge of the board material to obtain image data under the corresponding lighting mode. a digital image processor, communicatively connected to the controller and the detection device, for generating control instructions to the controller and receiving image data generated by the detection device and performing edge defect analysis based on the received image data, The detection device includes a lens imaging device, a mirror imaging device and a machine frame, the machine frame is a one-way opening structure, the opening end of the machine frame is arranged relative to the edge of the board material, the mirror imaging device and the lens imaging device are arranged relatively in the machine frame, the mirror imaging device is used to form a virtual image of the upper edge and the lower edge of the board material, and the lens imaging device is used to obtain a real image of the edge of the board material and a virtual image formed in the mirror imaging device; The mirror imaging device is located at the closed end of the frame and is arranged at an angle, and the lens imaging device is located at the open end of the frame and is located above or below the frame and is located at the angle bisector of the mirror imaging device, so that the distance between the real image of the lens imaging device and the edge of the panel material and the virtual image formed in the mirror imaging device is equal; Also includes: Distance sensor, used to measure the distance between the edge of the board material and the reference object. Speed sensor, used to measure the moving speed of the board material. A pan / tilt device, used to adjust the angle of the detection device relative to the edge of the panel material and / or to maintain the distance between the detection device and the edge of the panel material, The distance measuring sensor and the speed measuring sensor are communicatively connected to the digital image processor, and the pan-tilt device is communicatively connected to the controller. The digital image processor generates a control instruction based on the detection results of the distance measuring sensor and the speed measuring sensor and transmits it to the controller. The controller controls the pan-tilt device to dynamically maintain the distance between the detection device and the edge of the board material according to the received control instruction. The digital image processor can also analyze the clarity of the acquired image and generate an angle adjustment control instruction for the detection device to the controller. The controller controls the pan-tilt device to correspondingly adjust the angle of the detection device relative to the edge of the board material according to the received control instruction, so that the lens imaging device can obtain clear real and virtual images.
2. The panel material edge defect detection system according to claim 1, characterized in that: The lens imaging device and the mirror imaging device are both provided with a gimbal and are in communication connection with a controller, and the controller controls the gimbal to adjust the angle of the lens imaging device and / or the mirror imaging device.
3. The panel material edge defect detection system according to claim 1, characterized in that: The number of the detection devices is multiple sets, and they are respectively arranged corresponding to multiple edges of the board material.
4. The panel material edge defect detection system according to claim 1, characterized in that: The lighting device is arranged in the frame.
5. The panel material edge defect detection system according to claim 1, characterized in that: The lighting device is a combined light source including a plurality of light sources. Under the control of the controller, the combined light source provides a plurality of lighting modes with different lighting brightness by time-sharing switching lighting between different combinations of the light sources.
6. The panel material edge defect detection system according to claim 1, characterized in that: The digital image processor comprises: The defect recognition module is used to obtain the image data of each channel through a segmentation algorithm and reorganize it according to the imaging angle to obtain independent images of each angle, and perform pattern matching with the image of the standard panel material to identify defects.
7. The panel material edge defect detection system according to claim 6, characterized in that: The digital image processor further includes at least one of the following: The quality grade classification module is used to collect statistics on defect analysis results to classify product quality and obtain quality grade information; A defect position analysis module is used to determine the length, width, area and three-dimensional relative position information of each defect according to the defect analysis results, and generate a plane distribution map and / or a three-dimensional distribution map of the plate type defects; The storage module is used to receive and store the defect analysis results.
Citation Information
Patent Citations
Panel material full-view surface detect detecting system
CN108872246A
Device and method for detecting cracking edge of front side surface of lens based on boat-shaped light source
CN108872263A
Process and apparatus for the inspection of objects, particularly bottles
CN1129803A
Detection apparatus
CN207816852U
Plate surface material edge defect detection system
CN212031322U
