Method, device and system for detecting blockage and loop rising in strip steel transportation and medium

Abstract

The invention belongs to the technical field of unmanned detection of strip steel transportation, and discloses a jamming and rising detection method, device and system for strip steel transportation and a medium, the detection method comprises the following steps: collecting a field image in real time, and extracting a strip side face mask of strip steel from the field image; identifying the conveyor belt according to the calibration parameters and generating an out-of-bound detection frame on the conveyor belt; outputting an out-of-bound strip steel mask by combining the strip side face mask and the out-of-bound detection frame; calculating a contour point set in the out-of-bound strip steel mask and outputting an out-of-bound strip steel contour; the outline of the out-of-bound strip steel is compared with a preset value to determine whether the strip steel is lagged or not, unmanned monitoring of strip steel transportation can be achieved, and whether the arching height reaches the lagging or not can be accurately judged when the strip steel arches.

Description

Technical Field

[0001] The present invention relates to the technical field of unmanned detection of strip steel transportation, and in particular to a method, equipment, system and medium for detecting jamming and unloading of strip steel transportation. Background Art

[0002] Due to the thinness of the strip in the conveying equipment, the steel strip head is prone to jamming, blocking or even lifting (arching and deformation of the strip) during the platform threading process. The average transmission speed of the strip is above 2m / s, and the lifting arch will be formed within the jamming time of 1s, which will lead to an unsmooth threading process. Manual intervention is required to ensure the smooth threading of the strip head, which slows down the strip transportation speed and reduces the efficiency. The large stress of the strip at the deformed part may cause quality defects of the strip and cause certain production safety hazards on site. Therefore, it is necessary to arrange manual observation and detection on site in real time to see if the strip is stuck or lifted. The height at which the strip is stuck and lifted is a key indicator affecting the transportation of the strip. Manual observation of the strip lifting height is affected by subjective consciousness and has low accuracy. It is also not conducive to the personal safety of the observer. With the development of deep learning models and machine vision technology, it is possible to replace manual monitoring of the strip jamming and lifting phenomenon on site. Summary of the invention

[0003] In order to solve the problem that manual observation of strip steel lifting height is not accurate enough, the present invention provides a method, device, system and medium for detecting jamming of strip steel transportation.

[0004] In order to solve the above technical problems, the present invention provides a method for detecting jams and loosening of strip steel transportation, the detection method comprising: real-time acquisition of on-site images, and extraction of strip side masks of the strip steel in the on-site images; identification of the conveyor belt according to calibration parameters and generation of an out-of-bounds detection frame on the conveyor belt; outputting an out-of-bounds strip steel mask in combination with the strip side mask and the out-of-bounds detection frame; calculating a contour point set in the out-of-bounds strip steel mask and outputting an out-of-bounds strip steel contour; and comparing the out-of-bounds strip steel contour with a preset value to determine whether the strip steel has jammed and loosened.

[0005] In an embodiment of the present invention, in the step of comparing the out-of-bounds strip profile with a preset value to determine whether the strip is stuck or not, it further includes: when the out-of-bounds strip profile is smaller than the preset value, determining that the strip is not currently stuck or not; when the out-of-bounds strip profile is larger than the preset value, determining that the strip is currently stuck or not.

[0006] In an embodiment of the present invention, after the step of determining that the strip is currently stuck and starting, it further includes: using a bounding box algorithm to calculate the distance between the upper and lower ends of the strip contour to obtain the bounding box height; using a bounding box algorithm to calculate the distance between the left and right ends of the strip contour to obtain the bounding box width; and proportionally transforming the bounding box height and the bounding box width to respectively obtain the strip starting height and the starting width.

[0007] In an embodiment of the present invention, the step of combining the side mask and the out-of-bounds detection frame to output the out-of-bounds strip mask further includes: using a mask algorithm to calculate the overlapping area of ​​the side mask and the out-of-bounds detection frame; marking the side mask within the overlapping area as the in-bounds strip mask; marking the side mask outside the overlapping area as the out-of-bounds strip mask; and outputting the out-of-bounds strip mask as the out-of-bounds strip mask.

[0008] In an embodiment of the present invention, the step of calculating the contour point set in the out-of-boundary steel strip mask and outputting the out-of-boundary steel strip contour further includes: using a contour search algorithm to calculate the out-of-boundary steel strip connected domain in the out-of-boundary steel strip mask; extracting boundary points in the out-of-boundary steel strip connected domain and obtaining the out-of-boundary contour point set; and outputting the out-of-boundary steel strip contour according to the out-of-boundary contour point set.

[0009] In an embodiment of the present invention, the step of extracting the strip side mask of the strip in the scene image further includes: extracting a preliminary strip side mask in the scene image according to the strip characteristics; using an image corrosion algorithm to eliminate the boundary points of the preliminary strip side mask, and outputting the strip side mask.

[0010] In an embodiment of the present invention, the out-of-bounds steel strip contour includes a point count threshold and / or an area threshold of the contour.

[0011] In order to solve the above technical problems, the present invention also provides a jamming and unbundling detection system for strip steel transportation, the detection system comprising: an image acquisition module, used to acquire on-site images in real time; a strip steel mask module, used to extract the strip side mask of the strip steel in the on-site image; a conveyor belt mask module, used to identify the conveyor belt according to calibration parameters and generate an out-of-bounds detection frame on the conveyor belt; a mask analysis module, used to output an out-of-bounds strip steel mask in combination with the strip side mask and the out-of-bounds detection frame; a point set calculation module, used to calculate the contour point set in the out-of-bounds strip steel mask and output the out-of-bounds strip steel contour; and a jamming and unbundling detection module, used to compare the out-of-bounds strip steel contour with a preset value to determine whether the strip steel has jammed and unbundled.

[0012] In order to solve the above technical problems, the present invention also provides a jamming and unbundling detection device for strip steel transportation, the detection device comprises a host computer and an image acquisition device, the image acquisition device is used to acquire on-site images in real time, and the host computer is configured to: extract the strip side mask of the strip steel in the on-site image; identify the conveyor belt according to calibration parameters and generate an out-of-bounds detection frame on the conveyor belt; output an out-of-bounds strip steel mask in combination with the strip side mask and the out-of-bounds detection frame; calculate the contour point set in the out-of-bounds strip steel mask and output the out-of-bounds strip steel contour; compare the out-of-bounds strip steel contour with a preset value to determine whether the strip steel has jammed and unbundled.

[0013] In order to solve the above technical problems, the present invention also provides a computer-readable storage medium, which stores one or more programs. One or more of the programs can be executed by one or more processors to implement the above detection method.

[0014] Compared with the prior art, the present invention has the following beneficial effects:

[0015] Through the technical solution of the present invention, unmanned monitoring of strip steel transportation can be achieved without arranging observers at the strip steel transportation site. At the same time, when the strip steel arches, it can be accurately judged whether the arch height reaches the jamming and starting standard, so that the operating personnel can adjust the strip steel conveying equipment according to the detection results, avoiding adjusting the strip steel conveying equipment without adjusting the strip steel arch height and affecting the strip steel transportation efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] The accompanying drawings are used to provide a further understanding of the embodiments of the present invention and constitute a part of the specification. Together with the following specific implementations, they are used to explain the embodiments of the present invention, but do not constitute a limitation on the embodiments of the present invention. In the accompanying drawings:

[0017] Figure 1 It is a flowchart of the steps of the method for detecting jamming and unloading of strip steel transportation provided by an embodiment of the present invention;

[0018] Figure 2 The overall logic diagram of the method for detecting jamming and unloading of strip steel transport provided by an embodiment of the present invention;

[0019] Figure 3 is a further flow chart of step S1, Figure 4 The image acquisition device captures the on-site image of the strip steel transportation when there is no steel stuck;

[0020] Figure 5 This is the effect diagram of detecting the side mask of the strip in the scene diagram of strip transportation when there is no steel jam;

[0021] Figure 6 The image acquisition device collects the on-site image of the strip transportation when the steel jam occurs;

[0022] Figure 7 It is the effect diagram of detecting the side mask in the strip transportation scene diagram when the steel jam occurs;

[0023] Figure 8 is a further flow chart in step S3;

[0024] Fig. 9 is a further flow chart in step S4; Fig.10 It is a rendering of the effect of detecting whether the strip side is out of bounds in the strip transportation scene diagram;

[0025] Fig.11 is a further flow chart in step S5;

[0026] Fig.12 is a further flow chart after step S501;

[0027] Fig.13 A module diagram for detecting jamming and unloading of strip steel transport provided in an embodiment of the present invention.

[0028] Description of Reference Numerals

[0029] 1. Detection system; 11. Image acquisition module; 12. Strip mask module; 13. Conveyor belt mask module; 14. Mask analysis module; 15. Point set calculation module; 16. Stuck steel detection module. DETAILED DESCRIPTION

[0030] The specific implementation of the present invention is described in detail below in conjunction with the accompanying drawings. It should be understood that the specific implementation described here is only used to illustrate and explain the present invention, and is not used to limit the present invention.

[0031] See also Figure 1 , Figure 1 It is a flow chart of the steps of the jamming detection method for strip steel transportation provided by an embodiment of the present invention. The semantic segmentation technology based on deep learning of the present invention adopts the EfficientNetV3 model as the neural network backbone, and the image of the strip steel is collected by an image acquisition device facing the conveyor belt. The data annotation scheme is to mark the edge of the strip steel on the side close to the image acquisition device as the side of the strip, and mark the flat area on the surface of the conveyor belt as the conveyor belt. The detection method of the present invention includes:

[0032] Step S1: collecting a live image in real time, and extracting a strip side mask of the strip from the live image;

[0033] Step S2: Identify the conveyor belt and generate an out-of-bounds detection frame on the conveyor belt according to the calibration parameters;

[0034] Step S3: combining the side mask and the out-of-bounds detection frame to output the out-of-bounds steel mask;

[0035] Step S4: Calculate the contour point set in the out-of-bounds steel mask and output the out-of-bounds steel contour;

[0036] Step S5: Compare the out-of-bounds strip profile with a preset value to determine whether the strip is stuck or not.

[0037] Specifically, the on-site image acquisition device first collects the on-site image of the conveyor belt area in real time. When there is strip steel on the conveyor belt, the strip steel can be identified based on the data annotation of the strip steel edge, and the side of the strip can be masked and tracked. At the same time, the conveyor belt is identified based on the data annotation of the conveyor belt, and an out-of-bounds detection frame is generated on the surface of the conveyor belt to detect whether the strip steel is out of bounds. When the strip steel does not get stuck or loose, the strip steel will move smoothly on the conveyor belt, and the mask of the side of the strip will be within the out-of-bounds detection frame throughout the whole process. When the strip steel gets stuck or loose, the strip steel will arch a certain distance on the conveyor belt. At this time, the extracted strip side mask will go beyond the out-of-bounds detection frame, and then combined with the out-of-bounds detection frame and the tracking of the strip side mask, the strip side mask that goes beyond the out-of-bounds detection frame is output, and this part of the mask is calibrated as the out-of-bounds strip steel mask, and the out-of-bounds strip steel contour is output by calculating the contour point set of the out-of-bounds strip steel mask and connecting the outermost points of the contour point set. Finally, by calculating the size of the out-of-bounds strip steel contour and comparing the calculated value with the preset value, it can be detected whether the strip steel is currently stuck or not, so that the operator can decide whether to adjust the conveying equipment according to the detection results.

[0038] The image acquisition device can be any of an industrial camera and a monitoring probe. The calibration parameters include the position coordinates of the image acquisition device relative to the conveyor belt. The preset value is a fixed value preset manually in the program, so as to compare the calculated out-of-bounds strip contour size with the preset value, thereby determining whether the strip is stuck or not. By extracting the strip side mask, the key features of the strip can be expressed more accurately, so as to facilitate the subsequent detection of the strip being stuck or not.

[0039] Through the strip steel jamming and unhooking detection method of the present invention, unmanned detection of the strip steel transported on the strip steel conveying equipment can be realized, avoiding potential safety hazards to the personnel who are specially observing the jammed steel at the work site. At the same time, when the strip steel is arched to a certain height, it can accurately detect whether the height of the strip steel arching reaches the jamming and unhooking standard, so that the operator can adjust the strip steel conveying equipment according to the detection result, avoiding adjusting the strip steel conveying equipment without adjusting the strip steel arching height and affecting the transportation efficiency of the strip steel. Through the actual application of transporting strip steel on multiple conveyor belts in a steel rolling mill, the accuracy rate of the present invention in detecting the strip side of the strip steel reaches 81.12%, and in the accuracy rate of judging the jamming and unhooking phenomenon of the strip steel, the false alarm rate reaches 100%, and the missed alarm rate reaches 100%.

[0040] See also Figure 2 , Figure 2 The overall logic diagram of the jam detection method for strip steel transportation provided by the embodiment of the present invention is provided to illustrate the overall inventive concept of the embodiment of the present invention. The rounded dotted rectangular container in the figure represents the input and output of the deep learning model or post-processing process, the rounded solid rectangular container represents the intermediate result of the algorithm, the rectangular solid container represents the processing method between the intermediate results or between the intermediate results and the input and output, and the diamond container represents the judgment and branch logic.

[0041] See also Figure 3-Figure 7 , Figure 3 is a further flow chart of step S1, Figure 4 The image acquisition device captures the scene picture of the strip steel transportation when there is no steel stuck. Figure 5 This is a picture showing the effect of detecting the side mask on the strip transportation site when there is no steel jam. Figure 6 The image acquisition device collects the on-site image of the strip transportation when the steel is stuck. Figure 7 It is a picture showing the effect of detecting the side mask in the strip transportation scene when the strip is stuck. Step S1 further includes the following steps:

[0042] Step S101: extracting a preliminary strip side mask from the live image according to the strip features;

[0043] Step S102: using an image erosion algorithm to eliminate the boundary points of the preliminary side mask, and outputting the side mask.

[0044] The preliminary strip side mask is corroded by the image corrosion algorithm to eliminate the boundary points in the preliminary strip side mask and obtain a strip side mask that shrinks inward and becomes thinner, so that the position information of the strip is expressed more accurately, thereby improving the accuracy of subsequent jam detection. The output strip side mask is as follows: Figure 5 and Figure 7 As shown in the figure, the thick line on the side of the strip close to the image detection device is the side mask of the strip.

[0045] See also Figure 8 , Figure 8 is a further flow chart of step S3. Step S3 further comprises the following steps:

[0046] Step S301: using a mask algorithm to calculate the overlapping area with the side mask and the out-of-bounds detection box;

[0047] Step S302: marking the strip side mask within the overlapping area as the in-bounds strip mask;

[0048] Step S303: marking the strip side mask outside the overlapping area as an out-of-bounds strip mask;

[0049] Step S304: Output the out-of-bounds steel strip mask as an out-of-bounds steel strip mask.

[0050] When the mask algorithm is used to calculate the overlapping area, a mask is generated based on the area enclosed by the out-of-bounds detection frame. The area where the side mask overlaps with the mask is the overlapping area. Then, using the overlapping area as a reference, the side mask within the overlapping area is marked as the in-bounds strip mask, and the side mask outside the overlapping area is marked as the out-of-bounds strip mask. The out-of-bounds strip mask is then output as the out-of-bounds strip mask. Similarly, the in-bounds strip mask can be output as the in-bounds strip mask, so that the out-of-bounds strip contour can be calculated based on the out-of-bounds strip mask.

[0051] See also Figure 9-10 , Fig. 9 is a further flow chart in step S4, Fig.10 It is a rendering of detecting whether the strip side is out of bounds in the strip steel transportation scene diagram. Step S4 further includes the following steps:

[0052] Step S401: using a contour search algorithm to calculate the connected domain of the out-of-bounds steel strip in the out-of-bounds steel strip mask;

[0053] Step S402: extracting boundary points in the connected domain of the out-of-bounds strip and obtaining an out-of-bounds contour point set;

[0054] Step S403: outputting the out-of-bounds strip steel contour according to the out-of-bounds contour point set.

[0055] The contour is found in the out-of-bounds steel strip mask by using the contour search algorithm and the connected domain of the out-of-bounds steel strip is calculated. Then, the boundary points are extracted from the connected domain of the out-of-bounds steel strip and an out-of-bounds contour point set is formed. The image in the out-of-bounds contour point set in the format of a two-dimensional grayscale numerical matrix is ​​converted into a sequence of two-dimensional discrete point sets to extract the out-of-bounds steel strip contour. Similarly, the connected domain of the in-bounds steel strip can be calculated by searching the contour in the in-bounds steel strip mask by using the contour search algorithm. The in-bounds steel strip contour can be obtained based on the connected domain of the in-bounds steel strip. Fig.10 As shown, the rectangle on the conveyor belt surface is the out-of-bounds detection frame, the contour line within the out-of-bounds detection frame is the in-bounds strip steel contour, and the contour line outside the out-of-bounds detection frame is the out-of-bounds strip steel contour.

[0056] in, Figure 4-Figure 7 as well as Fig.10 All images are decolorized and Figure 5 and Figure 7 In the figure, the extracted side mask is represented by the red area (thick line in the figure). Fig.10 In the figure, the out-of-bounds detection frame is represented by a green wireframe, the out-of-bounds strip outline is circled with a red thin strip, and the in-bounds strip outline is circled with a blue thin line, so that the operating personnel can remotely view the real-time situation of strip transportation through the display equipment.

[0057] See also Fig.11 , Fig.11 is a further flow chart of step S5. Step S5 further comprises the following steps:

[0058] Step S501: when the out-of-bounds strip profile is smaller than a preset value, it is determined that the strip is not currently stuck;

[0059] Step S502: When the out-of-bounds strip profile is greater than a preset value, it is determined that the strip is currently stuck.

[0060] After obtaining the out-of-bounds strip contour, the size of the out-of-bounds strip contour is calculated to determine whether the strip is currently experiencing a jam or loosening phenomenon. Specifically, the size of the out-of-bounds strip contour is compared with a preset value, wherein the out-of-bounds strip contour includes a point threshold and / or an area threshold of the contour. When either the point threshold and the area threshold is less than the preset value, it is determined that the strip is not currently experiencing a jam or loosening phenomenon. Similarly, when either the point threshold and the area threshold is greater than the preset value, it is determined that the strip is currently experiencing a jam or loosening phenomenon.

[0061] The preset value includes a preset point value and a preset area value. For example, the preset point value is set to 20. When the point threshold of the out-of-bounds strip steel contour is less than 20, it can be determined that the strip steel is not currently stuck, otherwise it is determined that it is stuck. For another example, the preset area value is set to 30mm 2 , when the area threshold of the out-of-bounds strip profile is less than 30mm 2 When the strip is detected, it can be determined that no jamming or loosening occurs at present; otherwise, it can be determined that jamming or loosening occurs.

[0062] See also Fig.12 , Fig.12 It is a further flow chart after step S501. Step S501 further includes the following steps:

[0063] Step S5011: using a bounding box algorithm to calculate the distance between the upper and lower ends of the out-of-bounds strip steel contour to obtain the bounding box height;

[0064] Step S5012: using a bounding box algorithm to calculate the distance between the left and right ends of the out-of-bounds strip steel contour to obtain the bounding box width;

[0065] Step S5013: Proportionally transform the height and width of the bounding box to obtain the starting height and starting width of the strip respectively.

[0066] When calculating the height of the bounding box, the highest point of the out-of-bounds strip contour arch is taken as the upper end, and the line of the out-of-bounds detection frame away from the image acquisition device is taken as the lower end. The height of the bounding box is obtained by calculating the vertical distance between the upper end and the lower end. When calculating the width of the bounding box, the intersection of the out-of-bounds strip contour and the out-of-bounds detection frame is taken as the left end, and the position where the strip passes through the conveyor belt is taken as the right end. The width of the bounding box is obtained by calculating the straight-line distance between the left end and the right end. Finally, according to the trigonometric function transformation, the height and width of the bounding box are proportionally changed to obtain the starting height and starting width of the strip, so that the operator can adjust the conveying equipment according to the starting height and starting width to reverse the steel, thereby removing the starting arch of the strip.

[0067] According to the above-mentioned method for detecting jamming and unloading of strip steel transportation, the present invention also provides a system 1 for detecting jamming and unloading of strip steel transportation, see Fig.13 , Fig.13 This is a module diagram of the jam detection of strip steel transportation provided by an embodiment of the present invention. The detection system 1 includes:

[0068] An image acquisition module 11 is used to acquire on-site images in real time;

[0069] The strip mask module 12 extracts the strip side mask of the strip in the live image;

[0070] A conveyor belt mask module 13, which identifies the conveyor belt according to the calibration parameters and generates an out-of-bounds detection box on the conveyor belt;

[0071] The mask analysis module 14 combines the strip side mask and the out-of-bounds detection frame to output the out-of-bounds strip steel mask;

[0072] Point set calculation module 15, calculates the contour point set in the out-of-bounds steel mask and outputs the out-of-bounds steel contour;

[0073] The steel jam detection module 16 compares the strip profile out of bounds with a preset value to determine whether the strip is jammed.

[0074] The image acquisition module 11 acquires the scene image in real time, and the strip mask module 12 extracts the strip side mask from the scene image, the conveyor belt mask module 13 generates an out-of-bounds detection frame on the conveyor belt, and then the mask analysis module 14 outputs the out-of-bounds strip mask according to the strip side mask and the out-of-bounds detection frame, and then the point set calculation module 15 calculates the out-of-bounds strip mask contour in the out-of-bounds strip mask, and finally the stuck steel detection module 16 determines whether the current strip is stuck or not. The entire detection process does not require manual participation, and there is no need to arrange observers at the strip transportation site. At the same time, when the strip is arched, it can also accurately determine whether the arch height reaches the jamming standard, so that the operator can adjust the strip conveying equipment according to the detection results, so as to avoid adjusting the strip conveying equipment without adjusting the strip arch height and affecting the transportation efficiency of the strip.

[0075] According to the above-mentioned loose coil abnormality detection method, an embodiment of the present invention further provides a jamming and unpacking detection device for strip steel transportation, the detection device includes a host computer and an image acquisition device, the image acquisition device is used to collect on-site images in real time, and the host computer is configured as follows:

[0076] Extracting the strip side mask of the strip in the live image;

[0077] Identify the conveyor belt according to the calibration parameters and generate an out-of-bounds detection box on the conveyor belt;

[0078] Combine the side mask and the out-of-bounds detection frame to output the out-of-bounds steel mask;

[0079] Calculate the contour point set in the out-of-bounds steel mask and output the out-of-bounds steel contour;

[0080] The out-of-bounds strip profile is compared with the preset value to determine whether the strip is stuck or not.

[0081] By using the detection equipment of the present invention to detect the jamming and lifting of the strip during transportation, an unmanned monitoring effect of the jamming and lifting of the strip can be achieved, without the need to arrange observers at the strip transportation site. At the same time, when the strip is arched, it can also accurately determine whether the arch height reaches the jamming and lifting standard, so that the operating personnel can adjust the strip conveying equipment according to the detection results, avoiding adjusting the strip conveying equipment without adjusting the strip arch height and affecting the transportation efficiency of the strip.

[0082] It should be understood that the specific features, operations and details described hereinabove with respect to the method of the present application may also be similarly applied to the device and system of the present application, or vice versa. In addition, each step of the method of the present application described above may be performed by a corresponding component or unit of the device or system of the present application.

[0083] It should be understood that each module / unit of the device of the present application can be implemented in whole or in part by software, hardware, firmware or a combination thereof. Each module / unit can be embedded in the processor of the electronic device or independent of the processor in the form of hardware or firmware, or can be stored in the memory of the electronic device in the form of software for the processor to call to perform the operation of each module / unit. Each module / unit can be implemented as an independent component or module, or two or more modules / units can be implemented as a single component or module.

[0084] According to the above-mentioned loose coil abnormality detection method, an embodiment of the present invention also provides a computer-readable storage medium, which stores one or more programs, and one or more of the programs can be executed by one or more processors to implement the above-mentioned strip transport jam detection method.

[0085] It will be appreciated by those skilled in the art that the method steps of the present invention may be performed by instructing related hardware such as an electronic device or a processor through a computer program, and the computer program may be stored in a non-temporary computer-readable storage medium, which causes the steps of the present application to be performed when the computer program is executed. Depending on the circumstances, any reference to memory, storage, or other media herein may include non-volatile or volatile memory. Examples of non-volatile memory include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM), flash memory, magnetic tape, floppy disk, magneto-optical data storage device, optical data storage device, hard disk, solid state disk, etc. Examples of volatile memory include random access memory (RAM), external cache memory, etc.

[0086] In the present invention, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.

[0087] The preferred embodiments of the present invention are described in detail above in conjunction with the accompanying drawings, but the present invention is not limited thereto. Within the technical concept of the present invention, the technical solution of the present invention can be subjected to a variety of simple modifications, including the combination of various specific technical features in any appropriate manner. In order to avoid unnecessary repetition, the present invention will not further describe various possible combinations. However, these simple modifications and combinations should also be regarded as the contents disclosed by the present invention and belong to the protection scope of the present invention.

[0088] Although the embodiments of the present invention have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present invention. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present invention.

Claims

1. A method for detecting jamming and unloading of strip steel transportation, characterized in that: The detection method comprises: Collect live images in real time and extract the strip side mask of the strip from the live images; Identify the conveyor belt according to the calibration parameters and generate an out-of-bounds detection box on the conveyor belt; Combine the side mask and the out-of-bounds detection frame to output the out-of-bounds steel mask; Calculate the contour point set in the out-of-bounds steel mask and output the out-of-bounds steel contour; The out-of-bounds strip profile is compared with the preset value to determine whether the strip is stuck or not.

2. The method for detecting jamming and unloading of strip steel transport according to claim 1, characterized in that: In the step of comparing the out-of-bounds strip profile with a preset value to determine whether the strip is stuck or not, the method further comprises: When the out-of-bounds strip profile is smaller than the preset value, it is determined that the strip is not currently stuck; When the out-of-bounds strip profile is greater than a preset value, it is determined that the strip is currently stuck.

3. The method for detecting jamming and unloading of strip steel transport according to claim 2, characterized in that: After the step of determining that the strip steel is currently stuck and starting, the method further includes: The bounding box algorithm is used to calculate the distance between the upper and lower ends of the out-of-bounds strip steel contour to obtain the bounding box height; The bounding box algorithm is used to calculate the distance between the left and right ends of the out-of-bounds strip steel contour to obtain the bounding box width; The height and width of the bounding box are proportionally transformed to obtain the starting height and starting width of the strip respectively.

4. The method for detecting jamming and unloading of strip steel transportation according to claim 1, characterized in that: In the step of combining the side mask and the out-of-bounds detection frame to output the out-of-bounds steel mask, further comprising: The masking algorithm is used to calculate the overlapping area with side masks and out-of-bounds detection boxes; The strip side mask within the marked overlapping area is the in-bounds strip mask; The side mask of the strip outside the marked overlapping area is the out-of-bounds strip mask; Output the out-of-bounds strip mask as out-of-bounds strip mask.

5. The method for detecting jamming and unloading of strip steel transportation according to claim 1, characterized in that: In the step of calculating the contour point set in the out-of-bounds steel mask and outputting the out-of-bounds steel contour, further comprising: The contour search algorithm is used to calculate the connected domain of the out-of-bounds steel strip in the out-of-bounds steel strip mask; Extract the boundary points in the connected domain of the out-of-bounds strip steel and obtain the out-of-bounds contour point set; Output the out-of-bounds strip steel contour based on the out-of-bounds contour point set.

6. The method for detecting jamming and unloading of strip steel transportation according to claim 1, characterized in that: The step of extracting the strip side mask of the strip in the scene image further includes: Extracting preliminary strip side masks in the live image based on strip features; An image erosion algorithm is used to eliminate the boundary points of the preliminary side mask and output the side mask.

7. The method for detecting jamming and unloading of strip steel transport according to claim 1, characterized in that: The out-of-bounds strip contour includes a point count threshold and / or an area threshold of the contour.

8. A jam detection system for strip steel transportation, characterized in that: The detection system comprises: Image acquisition module, used for real-time acquisition of on-site images; Strip mask module, extracts the strip side mask of the strip in the live image; The conveyor belt mask module identifies the conveyor belt according to the calibration parameters and generates an out-of-bounds detection box on the conveyor belt; The mask analysis module combines the side mask and the out-of-bounds detection box to output the out-of-bounds steel mask; A point set calculation module calculates the contour point set in the out-of-bounds steel mask and outputs the out-of-bounds steel contour; The stuck steel detection module compares the profile of the out-of-bounds strip steel with the preset value to determine whether the strip steel is stuck or not.

9. A jam detection device for strip steel transportation, characterized in that: The detection device includes a host computer and an image acquisition device, wherein the image acquisition device is used to acquire on-site images in real time, and the host computer is configured as follows: Extracting the strip side mask of the strip in the live image; Identify the conveyor belt according to the calibration parameters and generate an out-of-bounds detection box on the conveyor belt; Combine the side mask and the out-of-bounds detection frame to output the out-of-bounds steel mask; Calculate the contour point set in the out-of-bounds steel mask and output the out-of-bounds steel contour; The out-of-bounds strip profile is compared with the preset value to determine whether the strip is stuck or not.

10. A computer-readable storage medium, characterized in that: The computer-readable storage medium stores one or more programs, and the one or more programs can be executed by one or more processors to implement the detection method according to any one of claims 1 to 7.

Images