Self-adjusting production line based on visual recognition automatic double-row multi-direction distribution

Abstract

The application discloses a kind of self-regulating production line based on visual recognition automatic double-column multidirectional distribution, it is related to production line technical field, specifically disclose first direction conveying roller, second direction conveying belt group, emergency exit, visual shooting module and processing module, wherein processing module is used to generate production line driving strategy according to conveying state information and the information of first actual product volume on first sub-conveying line, and drive first direction conveying roller, second direction conveying belt group, lifting mechanism and baffle mechanism by production line driving strategy, the application can realize the proportional allocation of different application directions of product by the above technical scheme, not only high efficiency, and allocation proportion accuracy controllable.

Description

Technical Field

[0001] This invention relates to the field of production line technology, and in particular to a self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation. Background Technology

[0002] In modern manufacturing, to adapt to ever-changing market demands and improve production efficiency, companies are increasingly inclined to adopt advanced automation technologies and intelligent systems. Among these, the application of automated production lines is a key automation technology for enterprises.

[0003] Most of the automated production lines used in the current technology are single-row production lines. For production requirements with two product application destinations, manual sorting and allocation are required. This not only consumes manpower, but also easily leads to large deviations in the allocation ratio during the sorting process. Therefore, there is an urgent need for a production line that can automatically allocate products for production requirements with two product application destinations. Summary of the Invention

[0004] The purpose of this invention is to provide a production line capable of automatically allocating production requirements for products with two application destinations.

[0005] This invention discloses a self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation, comprising:

[0006] The first direction conveying roller conveyor includes several transfer rollers that rotate in the same direction, and one end of the first direction conveying roller conveyor is connected to the main conveyor line.

[0007] The second direction conveyor belt group includes several second direction conveyor belts. The second direction conveyor belts are arranged between the transfer rollers and are driven to move up and down by a lifting mechanism. One end of the second direction conveyor belt is connected to the first branch conveyor line and the other end is connected to the second branch conveyor line.

[0008] An emergency exit is provided, and a baffle mechanism is provided at the emergency exit. The emergency exit is connected to the end of the first direction conveyor roller.

[0009] The visual imaging module is used to capture the volume of the first actual product being transported on the first conveyor line;

[0010] The processing module is used to generate a production line drive strategy based on the conveying status information and the information on the first actual product volume on the first sub-conveyor line, and drive the first direction conveyor roller, the second direction conveyor belt group, the lifting mechanism and the baffle mechanism through the production line drive strategy.

[0011] In some embodiments disclosed in this invention, the production line driving strategy includes:

[0012] The conveying speed of the first direction conveyor roller, the conveying speed of the second direction conveyor belt group, the conveying direction of the second direction conveyor belt group, and the opening and closing action of the baffle mechanism in different time periods.

[0013] In some embodiments disclosed in this invention, the method for generating a production line driving strategy based on product delivery task information includes:

[0014] Analyze the product delivery task information to determine the total product volume delivered by the main delivery line in different time periods, the first preset product volume delivered by the first sub-delivery line in different time periods, and the second preset product volume delivered by the second sub-delivery line in different time periods.

[0015] Based on the total volume of products conveyed by the main conveyor line in different time periods, the conveying rate of the first conveyor roller in different time periods is determined.

[0016] Based on the first preset product volume conveyed by the first sub-conveyor line in different time periods and the second preset product volume conveyed by the second sub-conveyor line in different time periods, the conveying speed of the second direction conveyor belt group in different time periods is determined.

[0017] In some embodiments disclosed in this invention, the method for generating production line driving strategies further includes:

[0018] Establish a time reference line for transportation on the production line, and delineate several time intervals based on the production reference line;

[0019] The preset product volumes conveyed by the first and second sub-conveyors in different time segments are marked on the corresponding time segments on the time reference line.

[0020] The actual volume of the first product conveyed on the first sub-conveyor line, which is collected in real time by the visual imaging module, is mapped and compared on the time reference line.

[0021] When the conveying direction of the second direction conveyor belt group points to the first sub-conveyor line, if the cumulative value of the first actual product volume within a consecutive first preset number of time intervals is greater than the cumulative value of the first preset product volume, and the difference in the cumulative values ​​is greater than the first preset difference value, then the conveying direction of the second direction conveyor belt group is changed.

[0022] When the conveying direction of the second direction conveyor belt group points to the second sub-conveyor line, if the cumulative value of the second actual product volume within a consecutive second preset number of time intervals is less than the cumulative value of the first preset product volume, and the difference in the cumulative values ​​is less than the second preset difference value, then the conveying direction of the second direction conveyor belt group is changed.

[0023] In some embodiments disclosed in this invention, the method for generating a production line driving strategy based on conveying status information includes:

[0024] If the first or second sub-conveyor line experiences an abnormal conveying condition, the baffle mechanism will be activated, and the lifting mechanism will be activated to lower the second direction conveyor belt.

[0025] In some embodiments disclosed in this invention, the method for determining transmission status information includes:

[0026] Visual imaging modules are installed for both the first and second sub-conveyor lines;

[0027] Analyze the real-time images of the products being transported captured by the vision imaging module to determine the dynamic characteristics of the products on the conveyor line during the transport process.

[0028] Based on the dynamic characteristics of the products on the sub-conveyor lines during the conveying process, it is determined whether there are any abnormal conveying states on the sub-conveyor lines.

[0029] In some embodiments disclosed in this invention, the method for determining the dynamic characteristics of products on a conveyor line during the conveying process includes:

[0030] The product conveying images are arranged according to the time series, and the parts of the product conveying images that are mapped to the conveying lines are scanned and analyzed frame by frame to determine the outline boundary of the product in each frame of the product conveying image.

[0031] For each frame of the product delivery image, a number of boundary feature points are set according to a first preset interval on the contour boundary.

[0032] Scan continuous boundary feature points, classify boundary feature points whose distance is less than a second preset interval into the same boundary feature point group, and associate the boundary feature point group with the corresponding image mapping block;

[0033] Mark the direction reference line for the transport direction of the conveyor line on the product transport image, and use the direction reference line as the middle line to draw the direction boundary line at the third preset interval on both sides of the direction reference line, and identify the area between the direction boundary lines as the reasonable transport area.

[0034] Compare and analyze the boundary feature points in the boundary feature point group on each frame of the product delivery image with the directional boundary line to determine the number of first-interest feature points in the boundary feature point group that are located outside the reasonable delivery area.

[0035] If the number of first-focused feature points in a boundary feature point group is greater than the first preset number of anomalies, then the boundary feature point group is identified as a deviated boundary feature point group.

[0036] If a set of boundary feature points within a first preset number of consecutive occurrences appears continuously within a reasonable conveying area, then the sub-conveying line is considered to be in an abnormal conveying state.

[0037] In some embodiments disclosed in this invention, the method for determining the dynamic characteristics of products on a conveyor line during the conveying process further includes:

[0038] Mark the transport distance scale on the direction reference line in the product transport image, and set a distance scale line perpendicular to the direction reference line for each transport distance scale point;

[0039] Within the first preset time interval, the boundary feature point group that continuously contacts the distance scale line less than the preset scale line number is identified as the same boundary feature point group.

[0040] Based on the time nodes when the boundary feature point group comes into contact with lines at different distance scales, the moving speed of the boundary feature point group is determined. If the moving speed of the boundary feature point group is lower than the preset moving speed, it is determined that there is an abnormal conveying state in the sub-conveyor line.

[0041] In some embodiments disclosed in this invention, the method for determining the dynamic characteristics of products on a conveyor line during the conveying process further includes:

[0042] The number of feature points in each boundary feature point group is determined. If the number of feature points in a boundary feature point group is greater than the preset number of feature points, the boundary feature point group is identified as a stacked boundary feature point group.

[0043] If the number of consecutive stacking boundary feature point groups is greater than the preset number of stacking feature point groups, then the sub-conveyor line is considered to have an abnormal conveying state.

[0044] In some embodiments disclosed in this invention, the method for determining the contour boundary of a product on a product transport image includes:

[0045] The product was photographed on-site to obtain a product calibration image;

[0046] For the product mapping area in the product calibration diagram, mark several visual acquisition points and define the color of the visual acquisition points to obtain the first color value range of the visual acquisition points;

[0047] The first color change gradient within a preset range near each visual acquisition point is determined, and the first color change gradient is associated with the visual acquisition point;

[0048] When analyzing the product transport image, scan a number of first pixels on the product transport image that match the first color value range;

[0049] Based on the degree of consistency of the first color change gradient within a preset range near the first pixel, several first pixels are filtered to obtain several second pixels;

[0050] Scan blocks are dynamically generated on the product delivery image until a second pixel with a number greater than the preset number of pixels is mapped in the scan block. Then the current scan block is identified as a reference scan block.

[0051] Edge analysis technology is used to perform edge analysis on the product conveying image to identify several sets of edge lines that are connected end to end on the product conveying image.

[0052] If the reference scan block belongs to a mapping block of an edge line group, then the edge line within the edge line group is considered as the outline boundary of the product.

[0053] This invention discloses a self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation, relating to the field of production line technology. Specifically, it discloses a first-direction conveyor roller conveyor, a second-direction conveyor belt group, an emergency exit, a visual imaging module, and a processing module. The processing module is used to generate a production line driving strategy based on the conveying status information and the information of the first actual product volume on the first sub-conveyor line. The production line driving strategy drives the first-direction conveyor roller conveyor, the second-direction conveyor belt group, the lifting mechanism, and the baffle mechanism. Through the above technical solution, this invention can achieve proportional allocation of products to different application destinations, which is not only highly efficient but also has controllable accuracy in allocation ratio.

[0054] The technical solution of the present invention will be further described in detail below with reference to the accompanying drawings and embodiments. Attached Figure Description

[0055] Figure 1 This is a schematic diagram of the structure of a self-adjusting production line based on visual recognition for automatic double-row multi-directional allocation, as disclosed in some embodiments of the present invention.

[0056] Figure 2 This is a flowchart illustrating the steps of a method for generating a production line driving strategy disclosed in some embodiments of the present invention.

[0057] Figure Labels

[0058] 1. First direction conveyor roller; 2. Second direction conveyor belt; 3. Emergency exit; 4. Baffle mechanism. Detailed Implementation

[0059] The technical solution of the present invention will be further described below with reference to the accompanying drawings and embodiments.

[0060] The technical solution of the present invention will be clearly and completely described below with reference to the accompanying drawings and specific embodiments. It should be understood that the preferred embodiments described herein are only for illustration and explanation of the present invention and should not be construed as limiting the scope of protection of the present invention. Those skilled in the art can make some non-essential improvements and adjustments based on the following content of the present invention. In the present invention, unless otherwise expressly specified and limited, the technical terms used in the present invention should have the ordinary meaning understood by those skilled in the art.

[0061] This invention discloses a self-adjusting production line based on visual recognition for automatic double-row multi-directional allocation. (See reference...) Figure 1 It includes: a first-direction conveyor roller 1, two sets of second-direction conveyor belts, an emergency exit 3, a vision shooting module, and a processing module.

[0062] The first directional conveyor roller 1 includes several transfer rollers rotating in the same direction, and one end of the first directional conveyor roller 1 is connected to the main conveyor line; the second directional conveyor belt 2 group includes several second directional conveyor belts 2, which are arranged between the transfer rollers and driven to move up and down by a lifting mechanism, one end of the second directional conveyor belt 2 is connected to a first sub-conveyor line, and the other end is connected to a second sub-conveyor line; a baffle mechanism 4 is provided at the emergency exit 3, and the emergency exit 3 is connected to the end of the first directional conveyor roller 1; the visual imaging module is used to capture the first actual product volume transported on the first sub-conveyor line; the processing module is used to generate a production line driving strategy based on the conveying status information and the information of the first actual product volume on the first sub-conveyor line, and drive the first directional conveyor roller 1, the second directional conveyor belt 2 group, the lifting mechanism, and the baffle mechanism 4 through the production line driving strategy.

[0063] In some embodiments disclosed in this invention, the production line driving strategy includes: the conveying rate of the first direction conveyor roller in different time periods, the conveying rate of the second direction conveyor belt group, the conveying direction of the second direction conveyor belt group, and the opening and closing action of the baffle mechanism.

[0064] In some embodiments disclosed in this invention, see [link / reference]. Figure 2 Methods for generating production line drive strategies based on product delivery task information include:

[0065] Step S100: Analyze the product conveying task information to determine the total product volume conveyed by the main conveyor line in different time periods, the first preset product volume conveyed by the first sub-conveyor line in different time periods, and the second preset product volume conveyed by the second sub-conveyor line in different time periods.

[0066] Step S200: Based on the total volume of products conveyed by the main conveyor line in different time periods, determine the conveying rate of the first conveyor roller in different time periods.

[0067] Step S300: Based on the first preset product volume conveyed by the first sub-conveyor line in different time periods and the second preset product volume conveyed by the second sub-conveyor line in different time periods, determine the conveying speed of the second direction conveyor belt group in different time periods.

[0068] The above steps involve comprehensively analyzing and adjusting the conveyor speeds of different conveyor lines to adapt to changes in production tasks. This analysis and adjustment process helps optimize the operation of the production line, improve production efficiency, and ensure the smooth flow of products throughout the manufacturing process.

[0069] In some embodiments disclosed in this invention, the method for generating production line driving strategies further includes:

[0070] Step S400: Establish a time reference line for transportation on the production line, and delineate several time segments based on the production reference line.

[0071] Step S500: Mark the preset product volume conveyed by the first and second sub-conveyor lines in different time segments on the corresponding time segments on the time reference line.

[0072] Step S600: Map and compare the actual volume of the first product being transported on the first sub-conveyor line, which is collected in real time by the visual imaging module, on the time reference line.

[0073] In step S700, when the conveying direction of the second direction conveyor belt group points to the first sub-conveyor line, if the cumulative value of the first actual product volume within a consecutive first preset number of time intervals is greater than the cumulative value of the first preset product volume, and the difference in the cumulative values ​​is greater than the first preset difference value, then the conveying direction of the second direction conveyor belt group is changed.

[0074] In step S800, when the conveying direction of the second direction conveyor belt group points to the second sub-conveyor line, if the cumulative value of the second actual product volume within a consecutive second preset number of time intervals is less than the cumulative value of the first preset product volume, and the difference in the cumulative values ​​is less than the second preset difference value, then the conveying direction of the second direction conveyor belt group is changed.

[0075] In some embodiments disclosed in this invention, the method for generating a production line driving strategy based on conveying status information includes: if an abnormal conveying state occurs in the first sub-conveyor line or the second sub-conveyor line, the baffle mechanism is opened and the lifting mechanism is driven to lower the second direction conveyor belt.

[0076] In some embodiments disclosed in this invention, the method for determining transmission status information includes:

[0077] Step S900: Visual imaging modules are set up for both the first and second sub-conveyor lines.

[0078] In this step, both the first and second sub-conveyor lines are equipped with vision shooting modules, with corresponding camera equipment responsible for capturing real-time images of the product conveying process.

[0079] Step S1000: Analyze the product conveying images captured in real time by the vision imaging module to determine the dynamic characteristics of the products on the conveying line during the conveying process.

[0080] This step involves analyzing real-time images of the product transported by the vision imaging module. The aim is to determine the dynamic characteristics of the products on the conveyor line during transport, including information such as their outline and position.

[0081] Step S1100: Based on the dynamic characteristics of the products on the sub-conveyor lines during the conveying process, determine whether there is an abnormal conveying state on the sub-conveyor lines.

[0082] The goal of this step is to determine whether there are any abnormal conveying conditions on the sub-conveyor lines.

[0083] In some embodiments disclosed in this invention, the method for determining the dynamic characteristics of products on a conveyor line during the conveying process includes:

[0084] Step S1001: Arrange the product conveying images according to the time series, and scan and analyze the portion of the product conveying image that maps to the conveying line frame by frame to determine the outline boundary of the product on each frame of the product conveying image.

[0085] In this step, the product delivery images are arranged according to time sequence, and the product's outline boundary is scanned and analyzed frame by frame, providing basic data for subsequent analysis.

[0086] Step S1002: Set several boundary feature points on the contour boundary of each frame of the product transmission image according to the first preset interval.

[0087] In this step, boundary feature points are set on the contour boundaries of each frame of the product image according to a first preset interval in order to extract key features.

[0088] Step S1003: Scan the continuous boundary feature points, classify the boundary feature points whose distance is less than the second preset interval into the same boundary feature point group, and associate the boundary feature point group with the corresponding image mapping block.

[0089] Step S1004: Mark the transport direction reference line on the product transport image for the transport direction of the sub-transport line, and use the direction reference line as the middle line to draw the direction boundary line at the third preset interval on both sides of the direction reference line, and identify the area between the direction boundary lines as the reasonable transport area.

[0090] Step S1005: Compare and analyze the boundary feature points in the boundary feature point group on each frame of the product delivery image with the directional boundary line to determine the number of first interest feature points in the boundary feature point group that are located outside the reasonable delivery area.

[0091] Step S1006: If the number of first focus feature points in the boundary feature point group is greater than the first preset number of anomalies, then the boundary feature point group is identified as a deviated boundary feature point group.

[0092] Step S1007: If a set of deviation boundary feature points in the first preset consecutive number appears continuously in the reasonable conveying area, it is determined that there is an abnormal conveying state in the sub-conveying line.

[0093] In some embodiments disclosed in this invention, the method for determining the dynamic characteristics of products on a conveyor line during the conveying process further includes:

[0094] Step S1008: Mark the transport distance scale on the direction reference line in the product transport image, and set a distance scale line perpendicular to the direction reference line for each transport distance scale point.

[0095] Step S1009: Within the first preset time interval, the boundary feature point groups that have a number of consecutive contact distance scale lines less than the preset number of scale lines are identified as the same boundary feature point group.

[0096] Step S1010: Determine the moving speed of the boundary feature point group based on the time nodes when the boundary feature point group comes into contact with the lines at different distance scales. If the moving speed of the boundary feature point group is lower than the preset moving speed, it is determined that there is an abnormal conveying state in the sub-conveyor line.

[0097] In the above series of steps, through in-depth analysis of real-time images, the system is able to promptly identify and respond to any abnormal conveying conditions that may exist on the conveyor line.

[0098] In some embodiments disclosed in this invention, the method for determining the dynamic characteristics of products on a conveyor line during the conveying process further includes:

[0099] Step S1011: Determine the number of feature points in each boundary feature point group. If the number of feature points in a boundary feature point group is greater than the preset number of feature points, then the boundary feature point group is identified as a stacked boundary feature point group.

[0100] Step S1012: If the number of consecutive stacking boundary feature point groups is greater than the preset number of stacking feature point groups, then it is determined that there is an abnormal conveying state in the sub-conveyor line.

[0101] In some embodiments disclosed in this invention, the method for determining the contour boundary of a product on a product transport image includes:

[0102] Step S10011: Take photos of the product on-site to obtain a product calibration image.

[0103] Step S10012: Mark several visual acquisition points in the product mapping area of ​​the product calibration diagram, and define the color of the visual acquisition points to obtain the first color value range of the visual acquisition points.

[0104] Step S10013: Determine the first color change gradient within a preset range near each visual acquisition point, and associate the first color change gradient with the visual acquisition point.

[0105] In step S10014, when analyzing the product conveying image, a number of first pixels on the product conveying image that conform to the first color value range are scanned.

[0106] Step S10015: Based on the degree of consistency of the first color change gradient within a preset range near the first pixel, several first pixels are filtered to obtain several second pixels.

[0107] Step S10016: Dynamically generate scanning blocks on the product delivery image until a second pixel point with a preset number of pixels is mapped in the scanning block, then the current scanning block is identified as a reference scanning block.

[0108] Step S10017: Using edge analysis technology, perform edge analysis on the product conveying image to determine several sets of edge lines that are connected end to end on the product conveying image.

[0109] Step S10018: If the reference scan block belongs to a mapping block of an edge line group, then the edge line in the edge line group is identified as the outline boundary of the product.

[0110] In the above steps, the product is photographed on-site to obtain a product calibration image; visual acquisition points are marked on the product calibration image, and their color ranges are defined; the color change gradient within a preset range near the visual acquisition points is determined and associated with the visual acquisition points; pixels on the product delivery image that conform to the color value range are scanned; second pixels that meet the conditions are selected based on the degree of conformity of the color change gradient; scanning blocks are dynamically generated until a second pixel is mapped within the scanning block that is greater than the preset number of pixels, then the scanning block is identified as a reference scanning block; edge analysis technology is used to determine the edge line groups that connect end to end on the product delivery image; if the reference scanning block is mapped to an edge line group, then the edge line of that edge line group is identified as the outline boundary of the product.

[0111] These steps, by analyzing the product's visual characteristics, including color and shape, and using edge analysis techniques, can effectively determine the product's contour boundaries during transportation, providing a foundation for subsequent anomaly detection.

[0112] Through the above description of the embodiments, those skilled in the art can clearly understand that the present invention can be implemented in hardware or by means of software plus necessary general-purpose hardware platforms. Based on this understanding, the technical solution of the present invention can be embodied in the form of a software product, which can be stored in a non-volatile storage medium (such as a CD-ROM, USB flash drive, external hard drive, etc.) and includes several instructions to cause a computer device (such as a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments of the present invention.

[0113] This invention discloses a self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation, relating to the field of production line technology. Specifically, it discloses a first-direction conveyor roller conveyor, a second-direction conveyor belt group, an emergency exit, a visual imaging module, and a processing module. The processing module is used to generate a production line driving strategy based on the conveying status information and the information of the first actual product volume on the first sub-conveyor line. The production line driving strategy drives the first-direction conveyor roller conveyor, the second-direction conveyor belt group, the lifting mechanism, and the baffle mechanism. Through the above technical solution, this invention can achieve proportional allocation of products to different application destinations, which is not only highly efficient but also has controllable accuracy in allocation ratio.

[0114] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to preferred embodiments, those skilled in the art should understand that modifications or equivalent substitutions can still be made to the technical solutions of the present invention, and these modifications or equivalent substitutions cannot cause the modified technical solutions to deviate from the spirit and scope of the technical solutions of the present invention.

Claims

1. A self-adjusting production line based on visual recognition for automatic double-row multi-directional distribution, characterized in that, include: The first direction conveying roller conveyor includes several transfer rollers that rotate in the same direction, and one end of the first direction conveying roller conveyor is connected to the main conveyor line. The second direction conveyor belt group includes several second direction conveyor belts. The second direction conveyor belts are arranged between the transfer rollers and are driven to move up and down by a lifting mechanism. One end of the second direction conveyor belt is connected to the first branch conveyor line and the other end is connected to the second branch conveyor line. An emergency exit is provided, and a baffle mechanism is provided at the emergency exit. The emergency exit is connected to the end of the first direction conveyor roller. The visual imaging module is used to capture the volume of the first actual product being transported on the first conveyor line; The processing module is used to generate a production line drive strategy based on the conveying status information and the information of the first actual product volume on the first sub-conveyor line, and drive the first direction conveyor roller, the second direction conveyor belt group, the lifting mechanism and the baffle mechanism through the production line drive strategy. The information determining the delivery status includes: Visual imaging modules are installed for both the first and second sub-conveyor lines; Analyze the real-time images of the products being transported captured by the vision imaging module to determine the dynamic characteristics of the products on the conveyor line during the transport process. Based on the dynamic characteristics of the products on the sub-conveyor line during the conveying process, determine whether there is an abnormal conveying state on the sub-conveyor line; Among them, the dynamic characteristics of the products on the conveyor line during the conveying process are determined as follows: The product conveying images are arranged according to the time series, and the parts of the product conveying images that are mapped to the conveying lines are scanned and analyzed frame by frame to determine the outline boundary of the product in each frame of the product conveying image. For each frame of the product delivery image, a number of boundary feature points are set according to a first preset interval on the contour boundary. Scan continuous boundary feature points, classify boundary feature points whose distance is less than a second preset interval into the same boundary feature point group, and associate the boundary feature point group with the corresponding image mapping block; Mark the direction reference line for the transport direction of the conveyor line on the product transport image, and use the direction reference line as the middle line to draw the direction boundary line at the third preset interval on both sides of the direction reference line, and identify the area between the direction boundary lines as the reasonable transport area. Compare and analyze the boundary feature points in the boundary feature point group on each frame of the product delivery image with the directional boundary line to determine the number of first-interest feature points in the boundary feature point group that are located outside the reasonable delivery area. If the number of first-focused feature points in a boundary feature point group is greater than the first preset number of anomalies, then the boundary feature point group is identified as a deviated boundary feature point group. If a set of boundary feature points of deviation within a first preset number of consecutive occurrences appears continuously in the reasonable conveying area, it is determined that there is an abnormal conveying state in the sub-conveying line. Among these, determining the dynamic characteristics of products on the conveyor line during the conveying process also includes: Mark the transport distance scale on the direction reference line in the product transport image, and set a distance scale line perpendicular to the direction reference line for each transport distance scale point; Within the first preset time interval, the boundary feature point group that continuously contacts the distance scale line less than the preset scale line number is identified as the same boundary feature point group. Based on the time nodes when the boundary feature point group comes into contact with lines at different distance scales, the moving speed of the boundary feature point group is determined. If the moving speed of the boundary feature point group is lower than the preset moving speed, it is determined that there is an abnormal conveying state in the sub-conveyor line.

2. The self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation according to claim 1, characterized in that, The production line driving strategy includes: The conveying speed of the first direction conveyor roller, the conveying speed of the second direction conveyor belt group, the conveying direction of the second direction conveyor belt group, and the opening and closing action of the baffle mechanism in different time periods.

3. The self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation according to claim 1, characterized in that, The production line drive strategy generated based on product delivery task information includes: Analyze the product delivery task information to determine the total product volume delivered by the main delivery line in different time periods, the first preset product volume delivered by the first sub-delivery line in different time periods, and the second preset product volume delivered by the second sub-delivery line in different time periods. Based on the total volume of products conveyed by the main conveyor line in different time periods, the conveying rate of the first conveyor roller in different time periods is determined. Based on the first preset product volume conveyed by the first sub-conveyor line in different time periods and the second preset product volume conveyed by the second sub-conveyor line in different time periods, the conveying speed of the second direction conveyor belt group in different time periods is determined.

4. The self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation according to claim 3, characterized in that, The production line-driven strategy also includes: Establish a time reference line for transportation on the production line, and delineate several time intervals based on the production reference line; The preset product volumes conveyed by the first and second sub-conveyors in different time segments are marked on the corresponding time segments on the time reference line. The actual volume of the first product conveyed on the first sub-conveyor line, which is collected in real time by the visual imaging module, is mapped and compared on the time reference line. When the conveying direction of the second direction conveyor belt group is pointing towards the first sub-conveyor line, if the cumulative value of the first actual product volume within a consecutive first preset number of time intervals is greater than the cumulative value of the first preset product volume, and the difference in the cumulative values ​​is greater than the first preset difference value, then the conveying direction of the second direction conveyor belt group is changed.

5. A self-adjusting production line based on visual recognition for automatic double-row multi-directional allocation according to claim 1, characterized in that, The production line drive strategy generated based on conveying status information includes: If the first or second sub-conveyor line experiences an abnormal conveying condition, the baffle mechanism will be activated, and the lifting mechanism will be activated to lower the second direction conveyor belt.

6. The self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation according to claim 1, characterized in that, Determining the dynamic characteristics of products on the distribution conveyor line during the conveying process also includes: The number of feature points in each boundary feature point group is determined. If the number of feature points in a boundary feature point group is greater than the preset number of feature points, the boundary feature point group is identified as a stacked boundary feature point group. If the number of consecutive stacking boundary feature point groups exceeds the preset number of stacking feature point groups, then the sub-conveyor line is considered to have an abnormal conveying state.

7. A self-adjusting production line based on visual recognition for automatic dual-row multi-directional allocation according to claim 1, characterized in that, Determining the product outline boundary on the product delivery image includes: The product was photographed on-site to obtain a product calibration image; For the product mapping area in the product calibration diagram, mark several visual acquisition points and define the color of the visual acquisition points to obtain the first color value range of the visual acquisition points; The first color change gradient within a preset range near each visual acquisition point is determined, and the first color change gradient is associated with the visual acquisition point; When analyzing the product transport image, scan a number of first pixels on the product transport image that match the first color value range; Based on the degree of consistency of the first color change gradient within a preset range near the first pixel, several first pixels are filtered to obtain several second pixels; Scan blocks are dynamically generated on the product delivery image until a second pixel with a number greater than the preset number of pixels is mapped in the scan block. Then the current scan block is identified as a reference scan block. Edge analysis technology is used to perform edge analysis on the product conveying image to identify several sets of edge lines that are connected end to end on the product conveying image. If the reference scan block belongs to a mapping block of an edge line group, then the edge line within the edge line group is considered as the outline boundary of the product.

Images