Online detection system and method for dynamically capturing disordered plasterboard based on vision

Abstract

The invention discloses an online detection system for dynamically capturing a disordered plasterboard based on vision, and the system comprises a main control module which is used for processing video data; the binocular vision module is arranged above the operation area and right faces the operation area, and the dual-mode vision module is in communication connection with the main control module so as to transmit collected video data to the main control module; the invention also provides a detection method. The binocular vision module is used for collecting video data of the operation area and transmitting the video data to the main control module, the main control module divides lines and sets side lines of video pictures of the operation area according to functions, edge marking is conducted on boards in the lines to form a closed wireframe so as to capture the motion state of the boards, and therefore the motion state of the boards can be captured according to the position relation between the closed wireframe and the side lines. And according to the position relation between the closed wire frames, whether the disordered plate appears is judged, and unattended disordered plate online identification is realized.

Description

Technical Field

[0001] The present invention relates to the technical field of gypsum board production, and in particular to an online detection system and method for gypsum board disorder based on visual dynamic capture. Background Art

[0002] During the high-speed production of gypsum board, if the sheets are not cut between each group, the two sheets will stick together, causing the gypsum boards to collide with the equipment bracket during the horizontal movement, resulting in disordered sheets. The gypsum boards will also tilt during the movement, causing disordered sheets during the flipping process, further causing waste. If there is any place that is not cleaned properly after the disordered sheets, the next group of sheets will continue to be disordered, causing more damage. Therefore, solving the problem of disordered gypsum boards during high-speed production can reduce the failure and downtime caused by disordered sheets, which is of great significance.

[0003] At present, the process of turning over and conveying gypsum boards side by side in the secondary control production line is usually detected by manual supervision. Although manual supervision can complete the task of detecting disordered boards, the conveying process of the boards is relatively simple in the production process. Long-term manual observation is prone to fatigue, and it is easy to fail to detect small-scale disordered boards in the early stage, which can easily lead to disordered boards in multiple routes later, causing malfunctions and shutdowns, resulting in large losses.

[0004] Therefore, it is difficult to detect minor board outages in time through manual observation and detection, which may lead to multi-area board outages and system downtime. Summary of the Invention

[0005] The purpose of the present invention is to provide an online detection system and method for gypsum board disorder based on visual dynamic capture, so as to solve the technical problem in the prior art that manual observation and detection are difficult to detect small-scale disorder in time, resulting in multi-area disorder and shutdown.

[0006] In order to solve the above technical problems, the present invention specifically provides the following technical solutions:

[0007] An online detection system for gypsum board disorder based on visual dynamic capture, including:

[0008] The main control module is located in the main control room and is used to process video data;

[0009] A binocular vision module is arranged above the working area and facing the working area, and the dual-mode vision module establishes a communication connection with the main control module to transmit the video data collected by it to the main control module;

[0010] The main control module divides the video data transmitted by the binocular vision module into eight adjacent lines according to their functions, and sets edges between adjacent lines. The main control module marks the edges of the boards in the eight lines to form a closed wireframe and captures their motion status.

[0011] When the closed wireframes in each line touch the edge line after longitudinal transport or the angle between them and the edge line during transverse transport exceeds the threshold, the main control module outputs a slant signal; when any two closed wireframes in the same line overlap, the main control module outputs an overlap signal; when two closed wireframes in adjacent lines are misaligned during longitudinal synchronous transport, the main control module outputs an alignment signal; when an irregular closed wireframe appears in a single line, the main control module outputs a broken board signal to alert the monitoring personnel.

[0012] As a preferred solution of the present invention, the dual-mode vision module includes camera 1 and camera 2, and camera 1 and camera 2 are set at different heights to collect video data on the conveyor line from different angles.

[0013] As a preferred embodiment of the present invention, it also includes:

[0014] An alarm device is arranged in the working area, and the alarm device establishes a communication connection with the main control module;

[0015] When the main control module sends one or more signals selected from the group consisting of a walking signal, an overlapping signal, an unaligned signal, and a broken board signal, the alarm device generates an audible and visual alarm according to the received signal.

[0016] In order to solve the above technical problems, the present invention further provides the following technical solutions:

[0017] A detection method using the above-mentioned online detection system based on visual dynamic capture of gypsum board disorder includes the following steps:

[0018] Step 100: After using the binocular vision module to collect video data of the work area, the main control module divides the work area according to function based on the video data to obtain eight lines, and sets edge segmentation for adjacent parts of the eight lines to form a static reference image layer;

[0019] Step 200: After using the binocular vision module to collect video data of the work area, the main control module calculates the board edges in the work area based on the video data to obtain a closed wireframe describing the length and width specifications of the board. The main control module also captures the position changes of the board in real time to capture the motion state of the closed wireframe and form a dynamic motion picture layer.

[0020] Step 300: The main control module overlays the motion image layer with the reference image layer, obtains the motion state of the closed wireframe in the route, and determines the abnormal transmission route based on the positional relationship between the closed wireframe and the edge line, and the positional relationship between the closed wireframes.

[0021] Step 400: When the main control module determines that there is an abnormal transmission line, it immediately sends a signal to prompt the monitoring personnel.

[0022] As a preferred embodiment of the present invention, in step 100, the eight lines of the reference image layer are, from left to right, a feeding line, a left flipping line, a right flipping line, a left arrangement line, a right arrangement line, and a discharging line, and there are three discharging lines, which output three sheets side by side.

[0023] Among them, the plates on the feeding line can move longitudinally and move laterally to the left flipping line. The left flipping line can cooperate with the right flipping line to flip the plates from the left to the right, and the plates on the right flipping line can move laterally to the left arrangement line.

[0024] The plates on the left arrangement line can move laterally to the right arrangement line, the plates on the right arrangement line can move laterally to the discharge line, and the plates on the discharge line can move laterally to the right and longitudinally.

[0025] As a preferred solution of the present invention, in step 200, the motion picture layer is located above the reference picture layer, and the specific method of calculating and obtaining the closed wireframe of the motion picture layer based on the video data is:

[0026] Based on the acquisition frame rate of the binocular vision module, the calculation frequency of the main control module is set so that the calculation frequency is equal to the acquisition frame rate, and calculation is performed on each frame of the video;

[0027] The main control module performs real-time calculations on each frame of video data based on the calculation frequency, obtains the contour data of the plate in each frame of video, and constructs a closed wireframe describing the edge of the plate in each frame of video based on the contour data.

[0028] As a preferred solution of the present invention, in step 300, based on the positional relationship between the closed wireframe and the edge line, and the positional relationship between the closed wireframes, the specific method of determining the abnormal transmission route is as follows:

[0029] Based on the calculation frequency of the main control module, the closed wireframe of the plate edge in each frame of the video is imported into the main control module;

[0030] The main control module composites the closed wireframe with the reference picture layer, analyzes the relationship between the closed wireframe and the edge line of the reference picture layer, and analyzes the relationship between the closed wireframe and the closed wireframe, and determines whether there is a chaotic board in each frame of the video based on the relationship between the closed wireframe and the edge line, and the relationship between the closed wireframe and the closed wireframe.

[0031] As a preferred solution of the present invention, the specific conditions for determining whether there is a chaotic board in each frame of video are:

[0032] When the closed wireframe touches the edge line during the longitudinal movement along the line, it is determined to be deflected. When the closed wireframe touches the edge line during the horizontal movement perpendicular to the line, it is determined to be deflected when the angle exceeds the threshold.

[0033] Overlap is determined when one of the closed wireframes overlaps with another closed wireframe of the same or adjacent line during horizontal or vertical movement;

[0034] Misalignment is determined when a front-to-back position difference occurs between one closed wire frame and another closed wire frame of an adjacent line during movement;

[0035] A broken board is determined when irregular lines appear in one of the closed wireframes.

[0036] Compared with the prior art, the present invention has the following beneficial effects:

[0037] The present invention utilizes a binocular vision module to collect video data of the working area and transmits it to a main control module. The main control module divides the video screen of the working area into lines and sets edges according to the function, and marks the edges of the boards in the lines to form a closed wireframe to capture their motion state. Based on the positional relationship between the closed wireframe and the edgeline, as well as the positional relationship between the closed wireframe and the closed wireframe, it is determined whether disordered boards appear, thereby realizing unattended online identification of disordered boards. BRIEF DESCRIPTION OF THE DRAWINGS

[0038] To more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for the embodiments or the description of the prior art. Obviously, the drawings described below are merely exemplary, and those skilled in the art can, without inventive effort, derive other implementation drawings based on the provided drawings.

[0039] Figure 1 A schematic structural diagram of an online detection system for gypsum board irregularities based on visual dynamic capture provided by an embodiment of the present invention;

[0040] Figure 2 A schematic diagram of determining gypsum board irregularity according to an online detection method for gypsum board irregularity based on visual dynamic capture provided by an embodiment of the present invention.

[0041] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of the present invention.

[0042] like Figure 1 As shown, the present invention provides an online detection system for gypsum board disorder based on visual dynamic capture, comprising:

[0043] The main control module is located in the main control room and is used to process video data;

[0044] The binocular vision module is arranged above the working area and faces the working area. The dual-mode vision module establishes a communication connection with the main control module to transmit the video data collected by it to the main control module;

[0045] The main control module divides the video data in the working area into eight adjacent lines according to the function based on the video data transmitted by the binocular vision module, and sets the edges between the adjacent lines. The main control module marks the edges of the boards in the eight lines to form a closed wireframe and captures their motion status.

[0046] When the angle between the closed wireframes in each line and the edge line after longitudinal transport or the edge line after transverse transport exceeds the threshold, the main control module outputs a slant signal; when any two closed wireframes in the same line overlap, the main control module outputs an overlap signal; when two closed wireframes in adjacent lines are misaligned during longitudinal synchronous transport, the main control module outputs an alignment signal; when an irregular closed wireframe appears in a single line, the main control module outputs a broken board signal to alert the monitoring personnel.

[0047] The present invention utilizes a binocular vision module to collect video data of the working area and transmits it to a main control module. The main control module divides the video screen of the working area into eight lines according to the function and sets edge lines. The board in the eight lines is edge-marked to form a closed wireframe to capture its motion state. Based on the positional relationship between the closed wireframe and the edge line, as well as the positional relationship between the closed wireframe and the closed wireframe, it is determined whether disordered boards occur, thereby realizing unattended online detection of disordered boards.

[0048] Specifically, the video data collected by the binocular vision module is transmitted to the main control module. The main control module processes the video data and divides the part of the video screen covering the working area according to the function of the working area to form eight lines. When a plate appears in the line, the main control module performs edge processing on the part of the video screen that is different from the line to obtain a closed wireframe describing the edge of the plate.

[0049] When the plate is conveyed in each line, the closed wire frame changes its position in the video screen. Therefore, when the closed wire frame in each line touches the edge line after longitudinal conveyance or the angle between the closed wire frame and the edge line during transverse conveyance exceeds the threshold, the main control module outputs a slant signal; when any two closed wire frames in the same line overlap, the main control module outputs an overlap signal; when two closed wire frames in adjacent lines are longitudinally conveyed synchronously and are out of position front to back, the main control module outputs an alignment signal; when an irregular closed wire frame appears in a single line, the main control module outputs a broken plate signal to alert the monitoring personnel.

[0050] Among them, the main control module needs to collect video data of empty plate operations when there is no plate in the operation area, obtain video data of the operation area, and then based on the processing of video data, use the inactive screen of the active operation area as the basis, and divide and set the edge lines on this basis to form various lines.

[0051] In the above-mentioned embodiment, the use of a binocular vision module for video data acquisition enables each video frame captured at the same time to have two different perspectives. Based on these two perspectives, the position information of the board can be calculated, thereby obtaining a closed wireframe and motion status. Because board irregularities can result in various problems, such as slanting, overlapping, misaligned, and broken boards, the following embodiment is provided to improve the accuracy of the binocular vision module in detecting irregular boards.

[0052] like Figure 1 As shown, the dual-mode vision module includes camera 1 and camera 2, and camera 1 and camera 2 are set high and low to collect video data on the conveyor line from different angles.

[0053] Specifically, the dual-mode vision module consists of camera one and camera two, and camera one is arranged at a high position and camera two is arranged at a low position, forming a height difference, so that camera one can collect video images of the plate surface from a high position, while camera two can collect the plate edge thickness from a low position. Based on the plate surface and plate edge thickness, it can more accurately judge the overlap and broken plates in the messy plates.

[0054] Of course, after the detection and determination of disordered boards, it is necessary to remind the staff in time, such as Figure 1 As shown, the online detection system also includes:

[0055] An alarm device is arranged in the working area and establishes a communication connection with the main control module;

[0056] Among them, when the main control module sends one or more signals among the deviation signal, overlap signal, misalignment signal and broken board signal, the alarm device will generate an audible and visual alarm according to the received signal.

[0057] Specifically, when the main control module detects that the boards in the circuit are out of order, it sends a signal. Upon receiving the signal, the alarm device emits an audible and visual alarm to alert staff. The alarm device can be a conventional warning light, or a combination of a display and warning light in the work area, providing an audible and visual alarm and displaying the detailed area of ​​the out-of-order boards.

[0058] Based on the above-mentioned online detection system for gypsum board disorder based on visual dynamic capture, a detection method is provided below, including the following steps:

[0059] Step 100: After using the binocular vision module to collect video data of the work area, the main control module divides the work area according to function based on the video data to obtain eight lines, and sets edge segmentation for adjacent parts of the eight lines to form a static reference image layer;

[0060] Step 200: After using the binocular vision module to collect video data of the work area, the main control module calculates the board edges in the work area based on the video data to obtain a closed wireframe describing the length and width specifications of the board. The main control module also captures the position changes of the board in real time to capture the motion state of the closed wireframe and form a dynamic motion picture layer.

[0061] Step 300: The main control module overlays the motion image layer with the reference image layer, obtains the motion state of the closed wireframe in the route, and determines the abnormal transmission route based on the positional relationship between the closed wireframe and the edge line, and the positional relationship between the closed wireframes.

[0062] Step 400: When the main control module determines that there is an abnormal transmission line, it immediately sends a signal to prompt the monitoring personnel.

[0063] The detection method of the present invention collects video data of the empty plate running in the working area through a binocular vision module, and obtains a video image of the conveying equipment with no change in the position of the working area, so that eight routes can be divided based on this, and functional zoning can be performed according to the video image to form a reference image layer with a border line. This reference image layer remains unchanged, which is convenient for comparison with the image of the plate movement, thereby providing recognition accuracy.

[0064] When the plates are transported in the line, the main control module processes the video images of the plates to obtain a closed wireframe describing the edge of the plates. The closed wireframe is moving relative to the edge line, and the closed wireframe is also located above the conveying equipment, that is, a moving picture layer covering the reference picture layer is formed. Through a composite analysis of the reference picture layer and the moving picture layer, the abnormal conveying line is determined based on the positional relationship between the closed wireframe and the edge line, as well as the positional relationship between the closed wireframe and the closed wireframe, so that a signal reminder can be sent immediately when an abnormal conveying line is identified.

[0065] Among them, in step 100, the eight lines of the reference picture layer are, from left to right, the feeding line, the left flipping line, the right flipping line, the left arrangement line, the right arrangement line and the discharging line, and there are three discharging lines, which output three plates side by side;

[0066] Among them, the plates on the feeding line can move longitudinally and move laterally to the left flipping line. The left flipping line can cooperate with the right flipping line to flip the plates from the left to the right, and the plates on the right flipping line can move laterally to the left arrangement line.

[0067] The plates on the left arrangement line can move laterally to the right arrangement line, the plates on the right arrangement line can move laterally to the discharge line, and the plates on the discharge line can move laterally to the right and longitudinally.

[0068] Specifically, there are eight parallel conveying lines in the working area, among which there is one feeding line and two flip lines, which are divided into a left flip line and a right flip line. After the feeding line feeds the gypsum board to its flip line, it can convey the gypsum board toward the flip line. The gypsum board conveyed to the left flip line can be moved to the right flip line by the flip.

[0069] There is an arrangement line on the other side of the flip line. There are two arrangement lines, namely the left arrangement line and the right arrangement line. The right flip line can transport the gypsum board on it toward the left arrangement line, and the left arrangement line can transport the gypsum board on it toward the right arrangement line to arrange the gypsum boards.

[0070] There are three discharging lines on the other side of the arrangement line. The discharging lines can transport the gypsum board both longitudinally and horizontally. That is, after the arrangement line on the right transports the gypsum board to the discharging line, the discharging line can transport the gypsum board to one of the discharging lines according to the set program.

[0071] In step 200, the motion picture layer is located above the reference picture layer, and the specific method of calculating and obtaining the closed wireframe of the motion picture layer according to the video data is as follows:

[0072] Based on the acquisition frame rate of the binocular vision module, the calculation frequency of the main control module is set so that the calculation frequency is equal to the acquisition frame rate, and calculation is performed on each frame of the video;

[0073] The main control module performs real-time calculations on each frame of video data based on the calculation frequency, obtains the contour data of the plate in each frame of video, and constructs a closed wireframe describing the edge of the plate in each frame of video based on the contour data.

[0074] Specifically, after the binocular vision module transmits the collected video data to the main control module, the main control module analyzes and calculates each frame of the video, so as to obtain the conveying equipment area when there is no board, that is, the working area, and form a reference picture layer based on this.

[0075] When a plate appears on each line, the main control module will obtain the plate's contour data through the data of the video images from two perspectives, and convert it into a closed wireframe of the plate. The closed wireframe is located above the reference image layer, and through frame-by-frame calculation and analysis, the position change of the closed wireframe in each frame can be obtained.

[0076] In step 300, based on the positional relationship between the closed wireframe and the edge line, and the positional relationship between the closed wireframes, the specific method for determining the abnormal transmission route is as follows:

[0077] Based on the calculation frequency of the main control module, the closed wireframe of the plate edge in each frame of the video is imported into the main control module;

[0078] The main control module composites the closed wireframe with the reference picture layer, analyzes the relationship between the closed wireframe and the edge line of the reference picture layer, and analyzes the relationship between the closed wireframe and the closed wireframe, and determines whether there is a chaotic board in each frame of the video based on the relationship between the closed wireframe and the edge line, and the relationship between the closed wireframe and the closed wireframe.

[0079] Specifically, the calculation program preset in the main control module pre-learns the relationship between the closed wireframes, side lines and closed wireframes of the edges of various chaotic boards. Therefore, after the video image of the board is input, it can automatically calculate and identify the positional relationship between the closed wireframe and the side lines, as well as the closed wireframe and the closed wireframe, so as to accurately determine whether a chaotic board appears in each frame of the video.

[0080] The specific conditions for determining whether there is a chaotic board in each frame of video are:

[0081] When the closed wireframe touches the edge line during the longitudinal movement along the line, it is determined to be deflected. When the closed wireframe touches the edge line during the horizontal movement perpendicular to the line, it is determined to be deflected when the angle exceeds the threshold.

[0082] Overlap is determined when one of the closed wireframes overlaps with another closed wireframe of the same or adjacent line during horizontal or vertical movement;

[0083] Misalignment is determined when a front-to-back position difference occurs between one closed wire frame and another closed wire frame of an adjacent line during movement;

[0084] A broken board is determined when irregular lines appear in one of the closed wireframes.

[0085] Specifically, the feeding line, the left flip line, the right flip line, the left arrangement line, the right arrangement line and the discharging line are set as L1 to L8 in sequence.

[0086] The deviation of L1 to L8 is divided into longitudinal deviation and transverse deviation. Among them, longitudinal deviation usually occurs between L1 and L6, L7 and L8, and transverse deviation usually occurs in the process of L1-L2, L3-L4-L5-L6-L7-L8.

[0087] Therefore, the determination of longitudinal inclination is as follows: Figure 2 As shown in a, for example, when there is a plate on L1 in the frame-by-frame analysis, and any area of ​​the closed wireframe obtained by the calculation touches the divided edge line, it is determined to be tilted. The judgment of horizontal tilt is as follows Figure 2 As shown in b, for example, when there is a plate on L1 in the frame-by-frame calculation and analysis and it moves laterally toward L2, if the inclination angle between the closed wireframe obtained by calculation and the edge of the plate is greater than the threshold, it is determined to be tilting.

[0088] The overlap judgment of L1 to L8 is divided into vertical overlap and horizontal overlap. Among them, vertical overlap usually occurs between L1 and L6, L7 and L8, and horizontal overlap usually occurs in the process of L3-L4-L5-L6-L7-L8.

[0089] Therefore, the determination of vertical overlap is as follows: Figure 2 As shown in d, for example, when there are adjacent plates on L1 calculated and analyzed frame by frame, and the edges of the plates are calculated and the two adjacent closed wireframes show a decrease in lines and area, it is determined to be overlapping. The determination of horizontal overlap is as follows Figure 2 As shown in d, for example, when there is a plate on L4 in the frame-by-frame analysis and it moves laterally toward L5 which also has a plate, when the closed wireframe in L4 and the closed wireframe in L5 decrease and the area decreases, it is determined to be overlapping.

[0090] The misalignment judgment of L1 to L8 usually occurs during the longitudinal movement and discharging of L6, L7, and L8. The judgment is as follows: Figure 2 As shown in c, for example, when there are adjacent plates in L6 and L7 calculated and analyzed frame by frame, and the position difference between the edges of the plate obtained by calculating the edges of two adjacent closed wire frames exceeds the threshold, it is determined to be misaligned.

[0091] For the judgment of broken boards in L1 to L8, it usually appears in L1 to L8. Therefore, the judgment of broken boards is as follows: Figure 2 As shown in FIG. 4 , for example, when there is a broken board on L1 calculated and analyzed frame by frame, if the closed wireframe specifications and area of ​​the broken board edge obtained by calculation are smaller than the closed wireframe specifications and area of ​​the intact board edge obtained by calculation, it is determined to be a broken board.

[0092] The above embodiments are merely exemplary embodiments of the present application and are not intended to limit the scope of the present application. The scope of protection of the present application is defined by the claims. Those skilled in the art may make various modifications or equivalent substitutions to the present application within the essence and scope of protection of the present application, and such modifications or equivalent substitutions shall also be deemed to fall within the scope of protection of the present application.

Claims

1. An online detection system for gypsum board disorder based on visual dynamic capture, characterized by: include: The main control module is located in the main control room and is used to process video data; A binocular vision module is arranged above the working area and facing the working area, and the dual-mode vision module establishes a communication connection with the main control module to transmit the video data collected by it to the main control module; The main control module divides the video data transmitted by the binocular vision module into eight adjacent lines according to their functions, and sets edges between adjacent lines. The main control module marks the edges of the boards in the eight lines to form a closed wireframe and captures their motion status. When the closed wireframes in each line touch the edge line after longitudinal transport or the angle between them and the edge line during transverse transport exceeds the threshold, the main control module outputs a slant signal; when any two closed wireframes in the same line overlap, the main control module outputs an overlap signal; when two closed wireframes in adjacent lines are misaligned during longitudinal synchronous transport, the main control module outputs an alignment signal; when an irregular closed wireframe appears in a single line, the main control module outputs a broken board signal to alert the monitoring personnel.

2. The online detection system for gypsum board disorder based on visual dynamic capture according to claim 1 is characterized in that: The dual-mode vision module includes camera 1 and camera 2, and camera 1 and camera 2 are arranged at different heights to collect video data on the conveyor line from different angles.

3. The online detection system for gypsum board disorder based on visual dynamic capture according to claim 2 is characterized in that: Also includes: An alarm device is arranged in the working area, and the alarm device establishes a communication connection with the main control module; When the main control module sends one or more signals selected from the group consisting of a walking signal, an overlapping signal, an unaligned signal, and a broken board signal, the alarm device generates an audible and visual alarm according to the received signal.

4. A detection method using the online detection system based on visual dynamic capture of gypsum board disorder according to any one of claims 1 to 3, characterized in that: The steps include: Step 100: After using the binocular vision module to collect video data of the work area, the main control module divides the work area according to function based on the video data to obtain eight lines, and sets edge segmentation for adjacent parts of the eight lines to form a static reference image layer; Step 200: After using the binocular vision module to collect video data of the work area, the main control module calculates the board edges in the work area based on the video data to obtain a closed wireframe describing the length and width specifications of the board. The main control module also captures the position changes of the board in real time to capture the motion state of the closed wireframe and form a dynamic motion picture layer. Step 300: The main control module overlays the motion image layer with the reference image layer, obtains the motion state of the closed wireframe in the route, and determines the abnormal transmission route based on the positional relationship between the closed wireframe and the edge line, and the positional relationship between the closed wireframes. Step 400: When the main control module determines that there is an abnormal transmission line, it immediately sends a signal to prompt the monitoring personnel.

5. A detection method according to claim 4, characterized in that, In step 100, the eight lines of the reference image layer are, from left to right, a feeding line, a left panel turning line, a right panel turning line, a left arrangement line, a right arrangement line, and a discharging line. There are three discharging lines, and three sheets of material are discharged side by side. Among them, the plates on the feeding line can move longitudinally and move laterally to the left flipping line. The left flipping line can cooperate with the right flipping line to flip the plates from the left to the right, and the plates on the right flipping line can move laterally to the left arrangement line. The plates on the left arrangement line can move laterally to the right arrangement line, the plates on the right arrangement line can move laterally to the discharge line, and the plates on the discharge line can move laterally to the right and longitudinally.

6. A detection method according to claim 5, characterized in that: In step 200, the motion picture layer is located above the reference picture layer, and the specific method of calculating and obtaining the closed wireframe of the motion picture layer according to the video data is as follows: Based on the acquisition frame rate of the binocular vision module, the calculation frequency of the main control module is set so that the calculation frequency is equal to the acquisition frame rate, and calculation is performed on each frame of the video; The main control module performs real-time calculations on each frame of video data based on the calculation frequency, obtains the contour data of the plate in each frame of video, and constructs a closed wireframe describing the edge of the plate in each frame of video based on the contour data.

7. A detection method according to claim 2, characterized in that, In step 300, based on the positional relationship between the closed wireframe and the edge line, and the positional relationship between the closed wireframes, the specific method of determining the abnormal transmission route is as follows: Based on the calculation frequency of the main control module, the closed wireframe of the plate edge in each frame of the video is imported into the main control module; The main control module composites the closed wireframe with the reference picture layer, analyzes the relationship between the closed wireframe and the edge line of the reference picture layer, and analyzes the relationship between the closed wireframe and the closed wireframe, and determines whether there is a chaotic board in each frame of the video based on the relationship between the closed wireframe and the edge line, and the relationship between the closed wireframe and the closed wireframe.

8. A detection method according to claim 2, characterized in that, The specific conditions for determining whether there is a chaotic board in each frame of video are: When the closed wireframe touches the edge line during the longitudinal movement along the line, it is determined to be deflected. When the closed wireframe touches the edge line during the horizontal movement perpendicular to the line, it is determined to be deflected when the angle exceeds the threshold. Overlap is determined when one of the closed wireframes overlaps with another closed wireframe of the same or adjacent line during horizontal or vertical movement; Misalignment is determined when a front-to-back position difference occurs between one closed wire frame and another closed wire frame of an adjacent line during movement; A broken board is determined when irregular lines appear in one of the closed wireframes.

Images