Liquid crystal panel conveying path intelligent control system based on visual identification

Abstract

The invention discloses a liquid crystal panel conveying path intelligent control system based on visual identification, and relates to the technical field of visual identification, and the system comprises the following steps: liquid crystal panel image identification is carried out to obtain panel space state information; performing spatial state judgment according to the panel spatial state information to obtain visual constraint information; performing path matching analysis according to the visual constraint information to obtain a path matching result; performing path feasibility judgment according to a path adaptation result to obtain a candidate conveying path set; performing path cost analysis according to the candidate conveying path set to obtain a target conveying path; and generating a control instruction according to the target conveying path to obtain a path control result. Through the step-by-step matching method of the constraint and the path structure, the conveying risk of the liquid crystal panel is reduced, a filtering result of a structure level is provided for subsequent steps, the stability and consistency of the whole conveying process are improved, the path decision complexity is reduced, and the path decision applicability is improved.

Description

Technical Field

[0001] This invention relates to the field of visual recognition technology, specifically to an intelligent control system for the conveying path of a liquid crystal panel based on visual recognition. Background Technology

[0002] LCD panels typically require frequent transport between multiple processes during manufacturing. Due to their large size, thinness, and limited rigidity, LCD panels demand high stability and shape integrity during transport. Any deviation in posture or abnormal shape during transport can easily lead to stress concentration at turning, support, or clamping points, causing panel damage and affecting product yield. With the development of computer vision technology, some transport systems have begun to incorporate visual recognition methods to detect the position or shape of LCD panels and control the transport accordingly.

[0003] In the existing technology, there are shortcomings in the matching analysis of the conveying path: existing LCD panel conveying systems mostly adopt preset conveying paths or path control methods based on fixed rules. That is, multiple conveying paths are planned in advance during the equipment design stage, and one of them is selected for conveying according to process requirements during operation. There is a lack of a mechanism to match and analyze the visual constraint information with the structural characteristics of the conveying path. It is impossible to effectively evaluate the structural adaptability of different conveying paths to the current panel state, thus affecting the applicability of path selection and the stability of conveying. Summary of the Invention

[0004] To address the shortcomings of existing technologies, this invention provides a visual recognition-based intelligent control system for the liquid crystal panel conveying path, thereby solving the problems mentioned in the background section.

[0005] To achieve the above objectives, the present invention provides the following technical solution: In a first aspect, the present invention provides an intelligent control system for the conveying path of a liquid crystal panel based on visual recognition, comprising the following steps: S1. LCD panel image recognition to obtain panel space status information; S2. Determine the spatial state based on the panel spatial state information to obtain visual constraint information; S3. Perform path matching analysis based on visual constraint information to obtain path adaptation results; S4. Determine the feasibility of the path based on the path adaptation results to obtain a set of candidate delivery paths; S5. Perform path cost analysis based on the candidate transport path set to obtain the target transport path; S6. Generate control commands based on the target transport path to obtain the path control results.

[0006] To further optimize this technical solution, the image recognition in step S1 includes: The conveying node is selected as the vision acquisition node. When the LCD panel passes through the node, the node image containing the overall outline of the LCD panel is acquired. The image is then analyzed and processed to identify the edge outline image of the LCD panel. The image coordinates are then mapped to the physical coordinate system of the conveying device to obtain the panel spatial state information at the conveying node.

[0007] To further optimize this technical solution, the spatial state determination in step S2 includes: Based on the obtained panel space state information, the position, orientation and shape of the LCD panel at the transport node are determined, clearly distinguishing between normal and abnormal states, and forming visual constraint information for transport path control.

[0008] To further optimize this technical solution, the path matching analysis in step S3 includes: Based on the obtained visual constraint information, the structural features of the candidate paths are matched and analyzed by a step-by-step matching method between constraints and path structures to identify the path that is structurally suitable for the current LCD panel and obtain the path adaptation result.

[0009] To further optimize this technical solution, the step-by-step matching method for constraints and path structures includes: Based on the position constraints in the visual constraint information, the carrying capacity of each conveying path is matched, and paths that do not meet the current position requirements of the panel are eliminated. On this basis, based on the attitude constraints, the turning structure of the remaining paths is matched, and paths that do not meet the attitude requirements of the panel are eliminated. Furthermore, based on the shape constraints, the support structure of the remaining paths is matched, and the path adaptation results are summarized.

[0010] To further optimize this technical solution, the path feasibility determination in step S4 includes: Based on the obtained path adaptation results and the transportation status information of the transportation path, the feasibility of the path is determined by a step-by-step judgment method with multiple conditions, based on the occupancy status and the stable operation status, to obtain a set of candidate transportation paths.

[0011] To further optimize this technical solution, the path cost analysis in step S5 includes: Based on the obtained set of candidate transport paths, the path cost value of each path is calculated and compared to achieve path cost analysis, thereby obtaining the target transport path that is most suitable for the current LCD panel transport.

[0012] To further optimize this technical solution, the path cost calculation includes:

[0013] in: : Conveying path The path value; : Conveying path Delivery time; Visual constraints affect the intensity; Time cost weight; : Conveying path The structural sensitivity coefficient; The path cost of the transport path is calculated by combining the transport time, path structure, and visual constraints.

[0014] To further optimize this technical solution, the control command generation in step S6 includes: Based on the obtained target conveying path, a conveying control command sequence corresponding to the path is generated to drive the conveying equipment to complete the path guidance of the LCD panel and obtain the path control result.

[0015] This technical solution has been further optimized, including the following functional modules: Panel image recognition module, visual constraint determination module, path structure matching module, path feasibility determination module, path cost calculation module, and transport control execution module.

[0016] In a second aspect, the present invention provides a computer device, including a memory and a processor, wherein the memory stores a computer program, wherein: when the computer program instructions are executed by the processor, they implement the steps of a vision recognition-based intelligent control system for liquid crystal panel transport paths as described in the first aspect of the present invention.

[0017] Thirdly, the present invention provides a computer-readable storage medium having a computer program stored thereon, wherein: when the computer program instructions are executed by a processor, they implement the steps of a vision recognition-based intelligent control system for liquid crystal panel transport paths as described in the first aspect of the present invention.

[0018] Compared with the prior art, the present invention provides an intelligent control system for the liquid crystal panel conveying path based on visual recognition, which has the following beneficial effects: This vision-based intelligent control system for LCD panel transport paths eliminates incompatible paths by using a step-by-step matching method between constraints and path structures. This prevents structurally mismatched paths from entering subsequent path selection stages, reducing the transport risk of LCD panels, providing structural-level filtering results for subsequent steps, improving the stability and consistency of the overall transport process, reducing the complexity of path decisions, and enhancing the applicability of path decisions. Attached Figure Description

[0019] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings used in the description of the embodiments will be briefly introduced below. Obviously, the accompanying drawings described below are only some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating an intelligent control system for the liquid crystal panel conveying path based on visual recognition proposed in this invention. Figure 2 This is a schematic diagram of a module of an intelligent control system for the conveying path of a liquid crystal panel based on visual recognition proposed in this invention. Detailed Implementation

[0021] To make the above-mentioned objects, features and advantages of the present invention more apparent and understandable, the specific embodiments of the present invention will be described in detail below with reference to the accompanying drawings.

[0022] Many specific details are set forth in the following description in order to provide a full understanding of the invention. However, the invention may also be practiced in other ways different from those described herein, and those skilled in the art can make similar extensions without departing from the spirit of the invention. Therefore, the invention is not limited to the specific embodiments disclosed below.

[0023] Secondly, the term "an embodiment" or "embodiment" as used herein refers to a specific feature, structure, or characteristic that may be included in at least one implementation of the present invention. The phrase "in one embodiment" appearing in different places throughout this specification does not necessarily refer to the same embodiment, nor is it a single embodiment or an embodiment selectively excluded from other embodiments.

[0024] Example 1: Reference Figure 1 This is the first embodiment of the present invention, which provides an intelligent control system for the liquid crystal panel conveying path based on visual recognition, including the following steps: S1. LCD panel image recognition to obtain panel space status information.

[0025] In this embodiment, the image recognition includes: During the transport of LCD panels, the structural complexity of the transport path is highly correlated with the physical state of the LCD panel itself. LCD panels are large, thin, and have limited rigidity, making them highly susceptible to non-ideal contact with the transport structure due to placement misalignment, angular deflection, or localized warping during transport. Relying solely on the operating signals or position switch status of the transport equipment cannot reflect the true spatial state of the LCD panel on the transport device, leading to a mismatch between the path selection and the actual state of the panel, resulting in jamming, collisions, or damage.

[0026] The transport node is selected as the vision acquisition node. When the LCD panel passes through the node, the node image containing the overall outline of the LCD panel is acquired. The image is analyzed and processed to identify the edge outline image of the LCD panel. The image coordinates are then mapped to the physical coordinate system of the transport device to obtain the panel spatial state information at the transport node. This transforms the LCD panel from a uniform processing object into a control object with state differences. This is the fundamental basis for all subsequent path control steps based on vision constraints and provides the input source for subsequent steps.

[0027] Image recognition steps include: Visual acquisition position setting: Select the conveying node that is crucial to the path diversion or path selection as the visual acquisition node. Set an industrial camera above or to the side of the node so that its shooting angle can completely cover the effective area of ​​the LCD panel, avoiding the loss of key edges due to obstruction. The visual acquisition position and the physical structure of the conveying device should maintain a fixed relative relationship to ensure that the mapping relationship between the subsequent image and the conveying coordinates is stable and reliable. Node image acquisition: When the LCD panel moves to the vision acquisition node with the conveying device, the industrial camera is triggered to acquire an image containing the overall outline of the LCD panel. This image is used to reflect the instantaneous placement state of the LCD panel at the current conveying node, giving the visual information direct control significance. Edge feature recognition: Image analysis and processing are performed on the acquired node images. Taking advantage of the regular shape, continuous edges and stable orientation of the liquid crystal panel, the main edge contours of the liquid crystal panel are identified. By the continuity and directionality of the edge features, the liquid crystal panel is distinguished from the background transport structure, thereby determining the effective area of ​​the liquid crystal panel in the image and providing basic data for subsequent spatial state calculation. Establishing coordinate correspondence: When the panel edge is in the image coordinate system, the unit is pixels, which is not directly comparable to the millimeter-level physical coordinates in the conveying system. Therefore, based on the pre-completed camera calibration results, the edge position in the image needs to be mapped to the physical coordinate system of the conveying device, so that the panel outline information originally in the image coordinates is converted into the position description in the conveying coordinates, so that the subsequent calculation results can be directly used to determine whether the panel deviates from the conveying center and whether there is a risk of attitude deviation. Spatial state information analysis: After coordinate mapping is completed, the spatial state information of the LCD panel at the conveying node is determined based on the distribution of the panel edge in the conveying coordinate system. This includes the overall position of the panel on the conveying device, the attitude deflection of the panel relative to the conveying direction, and whether there is asymmetry or abnormal trend in the distribution of the panel edge. This information serves as the final output of step S1 and provides basic input for subsequent steps.

[0028] S2. Determine the spatial state based on the panel spatial state information to obtain visual constraint information.

[0029] In this embodiment, the spatial state determination includes: In step S1, the spatial state information of the LCD panel at the transport node has been obtained through visual means. However, this spatial state information is still a raw state description, reflecting more the current position and orientation of the panel than addressing the constraints that this state imposes on the transport path. In actual transport, different transport paths place different requirements on the LCD panel. For example, paths with bending or clamping structures have a significantly lower tolerance for panel orientation deflection and warping than straight transport paths. If the spatial state information obtained in step S1 is directly used for path selection, the constraints imposed by these structural differences on the panel state cannot be reflected, potentially leading to insufficient basis for path control.

[0030] Based on the obtained panel spatial state information, the position, orientation, and shape of the LCD panel at the transport node are determined, clearly distinguishing between normal and abnormal states, forming visual constraint information for transport path control. This transforms the continuously changing spatial state into constraint information with clear control significance, reflecting the panel's adaptation requirements to the transport path structure, and providing a unified constraint basis for subsequent steps.

[0031] Methods for determining spatial state include: Panel spatial state information parsing: Receive the panel spatial state information output in step S1. This information includes the position description, attitude description and edge distribution of the panel at the conveying node. Parse the spatial state information and classify it into three categories: position-related information, angle-related information and edge shape-related information, in order to prepare for subsequent attitude determination and shape determination. Panel position status determination: During the LCD panel transport process, even if the overall posture and shape of the panel are normal, as long as its position on the transport device is significantly offset, it may make abnormal contact with baffles or support edges when entering certain path structures. Therefore, based on the description of the panel's position in the transport coordinate system, it is determined whether the panel is within the effective load-bearing area of ​​the transport device. By comparing the relative relationship between the center position of the panel and the center line of the transport path, it is determined whether there is a significant lateral offset of the panel. This is used to reflect whether the panel may make abnormal contact with the edge structure of the path during subsequent transport, thereby providing a position constraint basis for the selection of the path structure. Panel posture determination: During the conveying process, the LCD panel should ideally be basically aligned with the conveying direction. However, due to the effects of starting, braking or previous processes, the panel may deflect at a certain angle. In bending or clamping structures, this will significantly increase the risk of jamming and edge collision. Therefore, based on the posture description of the panel relative to the conveying direction, it is necessary to determine whether the panel has undergone a significant angle deflection. This determination does not focus on the absolute angle value of the panel, but rather on whether it exceeds the general tolerance range of the conveying path structure for posture deflection. This distinguishes between panels with normal posture and panels with a rotational tendency, providing a basis for determining whether to allow entry into bending structures or compact paths. Panel shape determination: During manufacturing and transportation, LCD panels may experience slight warping or local deformation due to uneven force, temperature changes, etc. These shape changes are often visually manifested as asymmetry or local deviation in edge distribution. If such conditions are ignored, the panel is prone to local force concentration and damage when entering a path with strong clamping force or small support spacing. Therefore, it is necessary to determine whether the panel has edge asymmetry or local shape changes based on the distribution of the panel edges in the transportation coordinate system. By analyzing the deviation of each edge of the panel relative to the reference plane, it is possible to identify the overall warping trend or local deformation trend. This reflects the degree of adaptation of the panel to the clamping and support structures during transportation and distinguishes between the two states of "stable shape" and "shape at risk". This is an important basis for determining whether the panel is suitable for entering a high-constraint path. Visual constraint information generation: The results of various judgments are summarized to form unified visual constraint information. This visual constraint information is used to describe the overall constraint requirements of the current LCD panel on the transport path structure at the transport node. It serves as the input condition for subsequent path adaptation analysis and path feasibility determination, ensuring that subsequent steps can perform path control under clear constraints.

[0032] S3. Perform path matching analysis based on visual constraint information to obtain path adaptation results.

[0033] In this embodiment, the path matching analysis includes: In actual LCD panel conveying systems, there are significant differences in structural form between different conveying paths, such as straight conveying paths, conveying paths with bends, and conveying paths with clamping or limiting structures. These structural differences directly determine the path's tolerance to the panel's posture, position, and shape. If the structural differences of the path are ignored during the path selection process, and judgment is made only based on equipment availability or conveying distance, some paths may be usable in operation but structurally unsuitable for the current panel state, or abnormal panels may be incorrectly assigned to high-constraint paths, thereby increasing the risk of jamming or damage.

[0034] Based on the obtained visual constraint information, the structural features of candidate paths are matched and analyzed using a step-by-step matching method between constraints and path structures. This identifies paths that are structurally suitable for the current LCD panel, yielding path adaptation results. Incompatible paths are then eliminated, preventing structurally mismatched paths from entering subsequent path selection stages. This reduces the transportation risk of the LCD panel, provides structural-level filtering results for subsequent steps, improves the stability and consistency of the overall transportation process, reduces the complexity of path decisions, and enhances the applicability of path decisions.

[0035] Furthermore, the step-by-step matching method between constraints and path structures includes: Based on the positional constraints in the visual constraint information, the load-bearing range of each conveying path is matched, and paths that do not meet the current position requirements of the panel are eliminated. On this basis, based on the attitude constraints, the turning structure of the remaining paths is matched, and paths that do not meet the attitude requirements of the panel are eliminated. Furthermore, based on the shape constraints, the support structure of the remaining paths is matched. The path adaptation results are then summarized. Specific implementation methods include: Visual constraint information reading: Receive and read the visual constraint information output in step S2, which describes the constraints imposed by the LCD panel on the transport path structure at the current position, including position constraints, attitude constraints and shape constraints, to ensure that the path matching analysis is based entirely on the current state of the panel; Path structure feature description acquisition: For each configured conveying path, read its corresponding structural feature description information to reflect the basic attributes of the path at the mechanical structure level, such as whether the path has a bending structure, the effective load-bearing width of the path, and whether the path contains clamping or limiting components, to ensure that the path matching analysis is based on the actual physical structural characteristics of the path. Path load-bearing range matching: Position constraints reflect the lateral placement of the LCD panel on the conveying device. However, the effective load-bearing range and edge structure of different paths differ. Based on the position constraints in the visual constraint information, it is determined whether the effective load-bearing range of each conveying path can accommodate the actual placement position of the current LCD panel. For paths with narrow load-bearing ranges or obvious edge limiting structures, if the panel position offset is close to or exceeds its structural allowable range, it is determined that the path is not suitable for the current panel in terms of position dimension, thereby eliminating paths that are sensitive to panel position and preventing abnormal contact of the panel in the early stage of entering the path. Path turning structure matching: Attitude constraints reflect the deflection state of the LCD panel relative to the transport direction. The bending, turning or attitude correction structures in the path have different degrees of sensitivity to the panel attitude. In the path through position matching, the attitude constraint information is further used to analyze whether the path contains bending, turning or attitude correction structures. For LCD panels with obvious attitude deflection, if the path contains continuous bending or structures with limited attitude adjustment capability, it is determined that the path is not suitable in the attitude dimension. For straight structures or paths with greater turning tolerance, they are retained as suitable paths. In this way, the panel attitude state is directly associated with the path turning structure, avoiding panels with abnormal attitude from entering high-risk paths with limited turning capability. Path support structure matching: Shape constraints reflect whether the LCD panel has a warping or local deformation trend. Different paths have significant differences in support continuity and clamping rigidity. In the path matching by position and posture, the support method and clamping strength of the path are further matched and analyzed based on the shape constraint information. For LCD panels with abnormal shape trends, if the path adopts a rigid clamping or a structure with a small support spacing, it is determined that the path is not suitable in shape dimension. For paths with continuous support and flexible clamping, it is determined to be a suitable path, thereby avoiding the abnormal shape panel from being subjected to additional stress in the high constraint structure and protecting the integrity of the panel structure. Path adaptation results are generated: After completing the structural matching analysis of position, attitude and shape, the matching results of each path are summarized to form path adaptation results. This identifies whether each path is suitable for the current LCD panel at the structural level, providing basic input for subsequent path feasibility determination by introducing the transport state.

[0036] S4. Determine the feasibility of the path based on the path adaptation results to obtain a set of candidate delivery paths.

[0037] In this embodiment, the path feasibility determination includes: In actual conveying systems, even if a path is structurally perfectly adapted to an LCD panel, it may still be unable to participate in the conveying process due to various factors during operation, such as path occupancy, operational conflicts, or unstable conditions. If path selection is made directly based solely on path adaptation results, it will lead to problems such as structurally adapted but currently occupied paths being mistakenly selected, paths in abnormal operating states entering the conveying process, or path structure judgments being mixed with operating condition judgments, making it difficult to pinpoint the source of the problem.

[0038] Based on the obtained path adaptation results and the transportation status information of the transportation path, a multi-state condition step-by-step judgment method is used to determine the feasibility of the path based on the occupied state and the stable operation state, thereby obtaining a set of candidate transportation paths. This clearly separates the structural judgment from the operational judgment, avoids the incorrect selection of paths that are structurally adapted but whose state is unavailable, reduces the judgment complexity of subsequent path selection steps, and improves the stability and interpretability of the path control process.

[0039] The steps for determining route feasibility include: Path adaptation result reading: Read the path adaptation result output in step S3. This result indicates whether each transport path is suitable for the current LCD panel at the structural level. By including only the structurally compatible paths in the subsequent analysis scope, all structurally incompatible paths no longer participate in any state judgment, thereby reducing the number of state analysis objects. Acquisition of conveying status information of conveying path: For the selected structurally compatible path, the corresponding conveying status information is acquired through the operation monitoring equipment of the conveying device. This information is used to reflect the current operating conditions of the path, such as whether the path is idle, whether there is a panel being conveyed, and whether the path is in a stable operating state. This ensures that the status information can reflect the real-time operation of the path, thereby avoiding reading the status of irrelevant paths, reducing the system load, and ensuring that the use of status information has a clear purpose. Path availability assessment based on occupancy status: In actual delivery tasks, the occupancy status of a path is the most direct and fundamental operational constraint. If a path is currently delivering other LCD panels, even if its structure is compatible and its operation is stable, it cannot simultaneously undertake new delivery tasks. Based on the occupancy status in the delivery status information, the availability of each structurally compatible path is assessed. For paths currently in an occupancy status, even if they are structurally perfectly compatible with LCD panels, they are not included in the candidate path range. For paths in an idle status, they are retained for the next step of assessment, thereby excluding paths that are temporarily unavailable due to parallel delivery or cycle time conflicts, and avoiding path conflicts and overlapping delivery tasks. Based on the stable operating state, the reliability of the path operation is judged: Even if the path is currently idle, if its operating state is unstable, such as being in the state switching, abnormal recovery or initialization stage, it is still not suitable to be used as a transportation path. If this factor is ignored, it may introduce unpredictable interruptions or abnormalities during the transportation process. Therefore, in the path judged by the occupancy status, the stable operating state in the transportation status information is further analyzed. For the path that is in the abnormal operation, state switching or has not completed initialization, it is determined that it does not have reliable transportation conditions at the current moment. For the path with a stable operating state, it is retained, thereby ensuring that the candidate path has continuous transportation capability at the operation level and avoiding the introduction of uncertain risks during the transportation process. Generate a candidate transport path set: After completing the step-by-step judgment of the occupied state and the stable operation state, the paths that meet all operating conditions are summarized to form a candidate transport path set, which is used to identify all paths that are both structurally suitable and operationally feasible at the current time, providing clear input for subsequent path selection or scheduling steps.

[0040] S5. Perform path cost analysis based on the candidate transport path set to obtain the target transport path.

[0041] In this embodiment, the path cost analysis includes: In LCD panel delivery tasks, if there is more than one candidate path and there is a lack of unified and quantitative selection criteria, problems such as path selection relying on fixed priorities, failing to reflect individual panel differences, and panels with abnormal posture or shape being assigned to paths with low structural tolerance may occur. Therefore, a path comparison step is needed to uniformly evaluate candidate paths.

[0042] Based on the obtained set of candidate transport paths, the path cost is calculated for each path and compared to achieve path cost analysis. The most suitable target transport path for the current LCD panel is obtained, thereby converging the candidate paths from "multiple available" to "one selected". Under the premise of ensuring transport feasibility, transport efficiency and panel safety are balanced.

[0043] Path cost analysis steps include: Candidate transport path set reading: Read the candidate transport path set output in step S4. Each path in the set has passed the structural adaptation judgment and operational feasibility judgment and has the basic conditions to be a target transport path, ensuring that subsequent cost calculation is only performed within the feasible path range. Transportation Path Information Acquisition: For each path in the candidate transportation path set, acquire its corresponding basic transportation attribute information to reflect the objective characteristics of the path in terms of transportation efficiency, including path length, estimated transportation time, and path structural complexity. For example, different path lengths directly affect transportation time and indirectly affect transportation efficiency, while structural complexity affects the stability during transportation. This information can be obtained through pre-configuration or real-time calculation, and its values ​​remain consistent within the same calculation cycle to ensure that the cost calculations between different paths are comparable and to avoid incomparable results between paths due to time changes or state disturbances. Cost influencing factors determination: The visual constraint information output in step S2 is introduced into the path cost calculation process. Based on the degree of attitude deflection and shape change of the LCD panel, the influence intensity of the visual constraint on the transport safety is determined. When the panel state is closer to the ideal state, the influence of the visual constraint on the path cost is smaller. When the panel has obvious deflection or abnormal shape trend, the influence of the visual constraint on the path cost is significantly enhanced. The influence intensity of the visual constraint is used as an adjustment factor to amplify or reduce the safety weight of different paths in the cost calculation, so that the path cost not only reflects the transport efficiency, but also reflects the degree of adaptation between the panel state and the path structure. Path cost calculation: For each path in the candidate transport path set, based on its basic transport attribute information and combined with the influence intensity of visual constraints, the corresponding path cost is calculated. Path length and transport time are used to reflect efficiency costs, while the influence intensity of visual constraints is used to reflect safety costs. For paths with low structural tolerance and high influence intensity of visual constraints, their cost will increase significantly, so that paths that are not conducive to the current panel state will be naturally separated in terms of numerical value. Target delivery path determination: After calculating the cost of all candidate paths, the cost of each path is compared, and the path with the lowest cost is selected as the target delivery path. Under the current state of the LCD panel, this target delivery path has both high delivery efficiency and low structural risk, and is the delivery path with the best overall cost.

[0044] Furthermore, the path cost calculation includes:

[0045] in: : Conveying path The path cost value is used to characterize candidate transport paths. The overall disadvantage of the current LCD panel condition is used as the sole comparison factor when selecting a path. It reflects the time cost of the path in terms of delivery efficiency and the sensitivity of the path to the visual constraints of the current panel in terms of structural stability. The smaller the value, the more suitable the path is to be selected under the current working conditions. : Conveying path The delivery time indicates the time it takes for the LCD panel to travel along the path. The estimated time required to transport from the current node to the target node can be calculated based on the path length and rated transport speed or obtained from historical statistics. The intensity of visual constraint influence reflects the impact of visual constraints on the path structure risk of the LCD panel. The more abnormal the panel's posture or shape, the larger the value. Time cost weight indicates the degree of attention paid to conveying efficiency during the panel conveying process. It is used to adjust the impact of conveying time on the overall cost and can be adjusted according to production cycle requirements. For example, the value is higher for high-cycle production lines and can be appropriately reduced for low-cycle or high-safety-requirement scenarios. : Conveying path The structural sensitivity coefficient represents the path The sensitivity to attitude deflection and shape change is used to amplify or reduce the impact of visual constraints on different paths. It reflects the number of turns, bending conditions, rigidity of clamping or limiting structures, and support continuity of the path. The more complex and rigid the path structure, the larger the value. It can be set according to equipment design parameters, engineering experience and historical data. For example, for straight and flexible support paths, a smaller value is usually taken, while for paths with many bends and strong clamping, a larger value is usually taken. The value should be consistent for similar paths. The path cost of the transport path is calculated by combining the transport time, path structure, and visual constraints.

[0046] Furthermore, the intensity of the visual constraint influence includes:

[0047] in: : The degree of attitude deflection indicates the degree of deflection of the LCD panel relative to the ideal transport posture, including in-plane rotation and tilt angle. It is used to reflect the intensity of the impact of abnormal LCD panel posture on transport safety. It is obtained from the posture determination result based on visual recognition in step S2 and can be scaled according to its allowable range. The value is close to 0 in the ideal posture. The more obvious the deflection, the larger the value. The degree of shape change indicates the level of shape change of the liquid crystal panel relative to the ideal flat state, such as the warping trend. It is used to reflect the degree of influence of abnormal liquid crystal panel shape on the transport stability. It is obtained from the posture determination result based on visual recognition in step S2 and can be scaled according to its allowable range. The value is smaller when the panel shape is normal and larger when the abnormal shape trend is more obvious. : Attitude weight coefficient, used to reflect the importance of attitude constraints. It can be adjusted according to panel type and process requirements. For processes that are more sensitive to attitude, the attitude weight coefficient is larger. The sum of the attitude weight coefficient and the shape weight coefficient is 1. Form weight coefficient: This coefficient reflects the importance of form constraints and can be adjusted according to panel type and process requirements. For processes with higher requirements for form integrity, the form weight coefficient is larger. The intensity of the visual constraint effect is obtained by weighting the degree of panel posture deflection and shape change.

[0048] Furthermore, the delivery time includes:

[0049] in: : Conveying path The path length is obtained based on the basic transport attribute information of the path; : Conveying path The conveying speed represents the speed at which items are conveyed along the conveying path, and can be obtained based on the basic conveying attribute information of the path or measurement data during stable operation. Based on the path length and conveying speed, the conveying time required for the panel to pass through the conveying path is calculated.

[0050] S6. Generate control commands based on the target transport path to obtain the path control results.

[0051] In this embodiment, the generation of control commands includes: In actual industrial conveying, the path selection result itself cannot directly drive equipment operation. Conveying equipment can only recognize and execute specific control commands, such as start, stop, turn, or switch. If the path control method only stays at the path selection level and lacks a clear mechanism for generating and issuing control commands, the path decision result cannot be directly used by the conveying system, resulting in a disconnect between visual guidance and actual conveying behavior, and making it difficult for the path control logic to form a complete engineering closed loop. Therefore, it is necessary to set up a dedicated execution step to map the target conveying path into explicit conveying control commands and guide the LCD panel along that path to complete the actual conveying.

[0052] Based on the obtained target conveying path, a sequence of conveying control instructions corresponding to the path is generated to drive the conveying equipment to complete the path guidance of the LCD panel, thereby obtaining the path control result. This transforms the target conveying path into executable equipment actions, realizing a closed-loop connection between the visual guidance result and the conveying control, ensuring that the LCD panel is conveyed strictly according to the calculated path, and improving the feasibility and reliability of the path control method in the industrial environment.

[0053] The steps for generating control commands include: Target transport path acquisition: Read the target transport path information output in step S5. This path information is used to describe the order of transport nodes and the corresponding transport channel relationship of the transport path that the LCD panel needs to pass through in the current transport task. The target transport path is a deterministic input and no path correction or re-judgment is performed to ensure that the execution stage strictly follows the decision results of the previous steps and prevents visual decision results from failing or transport behavior from being inconsistent due to path re-judgment. Conveying node sequence parsing: Based on the target conveying path information, the sequence of conveying nodes that the LCD panel needs to pass through in sequence during the conveying process is parsed out. Each conveying node corresponds to a specific conveying unit or transfer position. This node sequence clarifies the sequential relationship of the LCD panel in the conveying process, providing a structured basis for the generation of subsequent control commands. Conveyor control command sequence generation: Based on the parsed conveyor node sequence, Sequential Function Chart (SFC) control technology is used to map the conveyor node sequence to the corresponding conveyor control command sequence. In this process, each conveyor node corresponds to one or more conveyor action steps, and the control commands are arranged in sequence according to the node sequence to form a complete conveyor control command sequence. As a mature industrial control technology, Sequential Function Chart control technology can ensure that each conveyor action is executed in a predetermined order, and the next action is triggered only after the previous action is completed, thereby ensuring the continuity of the conveyor process. Drive the conveying equipment: Send the generated conveying control command sequence to the corresponding conveying equipment control unit, and the conveying equipment executes the corresponding conveying actions according to the command sequence, such as start, transfer or stop. During the execution, no adjustment is made to the target conveying path to ensure that the LCD panel moves strictly according to the path determined in step S5. Path guidance is completed: As the conveying control commands are executed in sequence, the LCD panel passes through each conveying node in sequence according to the target conveying path, and finally completes the conveying process from the starting position to the target position, realizing a complete closed loop from visual perception, path decision-making to conveying execution.

[0054] Example 2: Reference Figure 2 This is the second embodiment of the present invention, which provides a visual recognition-based intelligent control system for the liquid crystal panel conveying path, including the following functional modules: Panel image recognition module: During the LCD panel transport process, it acquires image information related to the transport node and outputs the spatial state information of the LCD panel based on the image recognition results. This information is used to characterize the position, orientation, and relative relationship of the LCD panel at the current transport node, providing basic input for subsequent visual constraint determination. Visual constraint determination module: Based on the panel space state information, it determines the posture and shape state of the LCD panel and outputs visual constraint information to reflect the degree of deviation of the LCD panel from the ideal delivery state. Path structure matching module: Analyzes the visual constraint information against the preset transport path structure characteristics, determines the degree of adaptation of different transport paths to the current LCD panel state at the structural level, and outputs the path adaptation result. Path Feasibility Determination Module: Based on the path adaptation results and combined with the current operating status information of the conveying system, the module determines the feasibility of each conveying path, selects a set of candidate conveying paths that meet the current conveying conditions, and provides the input range for path cost calculation. Path cost calculation module: Performs comprehensive cost calculation on each path in the candidate transport path set, determines the path with the minimum comprehensive cost based on transport time cost and structural constraint cost, and identifies this path as the target transport path for subsequent transport execution; Conveying control execution module: Converts the target conveying path into a corresponding conveying control command, and sends the control command to the conveying equipment to drive the LCD panel to complete the conveying process along the target conveying path, thereby realizing the actual execution of the path decision result.

[0055] Example 3: In practical applications, this invention can be applied to cross-process conveying path control in liquid crystal display panel production lines. It is used to determine the appropriate conveying path based on the real-time visual status of the liquid crystal panel in a multi-station, multi-path conveying environment. The typical scenario is the conveying process of a large-size liquid crystal panel between the cutting process and the bonding process.

[0056] In this practical application scenario, after the LCD panel completes the upstream cutting process, it is placed at the starting conveying node of the conveying system. The conveying system has multiple branch conveying paths arranged sequentially along the panel conveying direction. The different paths differ in the number of turns, bending radius, and support structure. Since the LCD panel may have different states such as posture deflection or edge warping after the cutting process, if it is directly conveyed according to the preset fixed path, interference risks may easily occur at the turning or limiting positions.

[0057] When the LCD panel enters the initial transport node, the transport system acquires a node image of the panel using a vision acquisition device located at that node. Based on this image, the system determines the LCD panel's positional relationship, orientation, and spatial state relative to the transport structure along the transport direction. Based on this spatial state information, the system further determines the degree of attitude deflection and morphological changes of the LCD panel, obtaining visual constraint information reflecting the current transport constraints of the panel.

[0058] After obtaining visual constraint information, the system analyzes its correspondence with the structural characteristics of multiple pre-set conveyor paths in the production line. Specifically, different conveyor paths differ in the number of turns, bending radius, and support continuity. Based on the visual constraint information, the system determines the degree of structural adaptability of each conveyor path to the current panel state and generates corresponding path adaptation results. Paths that are structurally unsuitable for the current panel posture or shape are marked as low-fit paths.

[0059] Based on the operational status information of the conveying equipment, the feasibility of paths with structural adaptability is determined, excluding paths that are occupied or temporarily unavailable for conveying, thus forming a set of candidate conveying paths. Within the candidate path range, the system comprehensively considers the conveying time corresponding to each path and the sensitivity of the path structure to the current visual constraints, calculates the comprehensive cost of each path, and determines the path with the minimum comprehensive cost as the target conveying path.

[0060] After the target conveying path is determined, the system generates conveying control commands according to the sequence of conveying nodes corresponding to that path, and sends the control commands to the conveying equipment control unit. The conveying equipment executes the corresponding conveying actions sequentially according to the control commands, causing the LCD panel to pass through each conveying node along the target conveying path, ultimately completing the conveying process from the cutting process to the bonding process.

[0061] Through the above specific embodiments, the present invention can dynamically determine the appropriate conveying path based on the real-time visual state of the LCD panel in the actual production environment, reduce the risk of interference or abnormal force on the LCD panel in the complex conveying structure, improve the stability and consistency of the conveying process, and is suitable for production line conveying application scenarios of large-size, high-precision LCD panels.

[0062] Example 4: This embodiment also provides a computer device applicable to a vision-based intelligent control system for the transport path of a liquid crystal panel, including a memory and a processor; the memory is used to store computer-executable instructions, and the processor is used to execute the computer-executable instructions to realize the vision-based intelligent control system for the transport path of a liquid crystal panel as proposed in the above embodiment.

[0063] This embodiment also provides a storage medium storing a computer program, which, when executed by a processor, implements a vision-based intelligent control system for the liquid crystal panel conveying path as proposed in the above embodiments.

[0064] The computer device can be a terminal, comprising a processor, memory, communication interface, display screen, and input devices connected via a system bus. The processor provides computing and control capabilities. The memory includes non-volatile storage media and internal memory. The non-volatile storage media stores the operating system and computer programs. The internal memory provides an environment for the operation of the operating system and computer programs stored in the non-volatile storage media. The communication interface is used for wired or wireless communication with external terminals; wireless communication can be achieved through Wi-Fi, carrier networks, NFC (Near Field Communication), or other technologies. The display screen can be an LCD screen or an e-ink screen. The input devices can be a touch layer covering the display screen, buttons, a trackball, or a touchpad on the computer device's casing, or an external keyboard, touchpad, or mouse.

[0065] If a function is implemented as a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods of the various embodiments of this invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0066] The logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be embodied in any computer-readable medium for use by, or in conjunction with, an instruction execution system, apparatus, or device (such as a computer-based system, a processor-including system, or other system that can fetch and execute instructions from, an instruction execution system, apparatus, or device). For the purposes of this specification, "computer-readable medium" can be any means that can contain, store, communicate, propagate, or transmit programs for use by, or in conjunction with, an instruction execution system, apparatus, or device.

[0067] More specific examples (a non-exhaustive list) of computer-readable media include: electrical connections (electronic devices) having one or more wires, portable computer disk drives (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable and editable read-only memory (EPROM or flash memory), fiber optic devices, and portable optical disc read-only memory (CDROM). Furthermore, computer-readable media can even be paper or other suitable media on which programs can be printed, because programs can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpreting, or otherwise processing as necessary, and then stored in computer memory.

[0068] It should be understood that various parts of the present invention can be implemented in hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented in software or firmware stored in memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented using any one or a combination of the following techniques known in the art: discrete logic circuits having logic gates for implementing logical functions on data signals, application-specific integrated circuits (ASICs) having suitable combinational logic gates, programmable gate arrays (PGAs), field-programmable gate arrays (FPGAs), etc.

[0069] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit it. 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 be made to the technical solutions of the present invention without departing from the spirit and scope of the technical solutions of the present invention, and all such modifications or substitutions should be covered within the scope of the claims of the present invention.

Claims

1. A visual recognition-based intelligent control system for the conveying path of a liquid crystal panel, characterized in that, Includes the following steps: S1. LCD panel image recognition to obtain panel space status information; S2. Determine the spatial state based on the panel spatial state information to obtain visual constraint information; S3. Perform path matching analysis based on visual constraint information to obtain path adaptation results; S4. Determine the feasibility of the path based on the path adaptation results to obtain a set of candidate delivery paths; S5. Perform path cost analysis based on the candidate transport path set to obtain the target transport path; S6. Generate control commands based on the target transport path to obtain the path control results.

2. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, The image recognition in step S1 includes: The conveying node is selected as the vision acquisition node. When the LCD panel passes through the node, the node image containing the overall outline of the LCD panel is acquired. The image is then analyzed and processed to identify the edge outline image of the LCD panel. The image coordinates are then mapped to the physical coordinate system of the conveying device to obtain the panel spatial state information at the conveying node.

3. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, The spatial state determination in step S2 includes: Based on the obtained panel space state information, the position, orientation and shape of the LCD panel at the transport node are determined, clearly distinguishing between normal and abnormal states, and forming visual constraint information for transport path control.

4. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, The path matching analysis in step S3 includes: Based on the obtained visual constraint information, the structural features of the candidate paths are matched and analyzed by a step-by-step matching method between constraints and path structures to identify the path that is structurally suitable for the current LCD panel and obtain the path adaptation result.

5. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 4, characterized in that, The step-by-step matching method between constraints and path structures includes: Based on the position constraints in the visual constraint information, the carrying capacity of each conveying path is matched, and paths that do not meet the current position requirements of the panel are eliminated. On this basis, based on the attitude constraints, the turning structure of the remaining paths is matched, and paths that do not meet the attitude requirements of the panel are eliminated. Furthermore, based on the shape constraints, the support structure of the remaining paths is matched, and the path adaptation results are summarized.

6. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, The path feasibility determination in step S4 includes: Based on the obtained path adaptation results and the transportation status information of the transportation path, the feasibility of the path is determined by a step-by-step judgment method with multiple conditions, based on the occupancy status and the stable operation status, to obtain a set of candidate transportation paths.

7. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, The path cost analysis in step S5 includes: Based on the obtained set of candidate transport paths, the path cost value of each path is calculated and compared to achieve path cost analysis, thereby obtaining the target transport path that is most suitable for the current LCD panel transport.

8. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 7, characterized in that, The path cost calculation includes: ； in: : Conveying path The path value; : Conveying path Delivery time; Visual constraints affect the intensity; Time cost weight; : Conveying path The structural sensitivity coefficient; The path cost of the transport path is calculated by combining the transport time, path structure, and visual constraints.

9. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, The generation of control commands in step S6 includes: Based on the obtained target conveying path, a conveying control command sequence corresponding to the path is generated to drive the conveying equipment to complete the path guidance of the LCD panel and obtain the path control result.

10. The intelligent control system for the liquid crystal panel conveying path based on visual recognition according to claim 1, characterized in that, Includes the following functional modules: Panel image recognition module, visual constraint determination module, path structure matching module, path feasibility determination module, path cost calculation module, and transport control execution module.

Images