CN-122018399-A - Liquid crystal display panel conveying path intelligent control system based on visual recognition

Abstract

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

Inventors

• YANG WEIPING
• LI CHAO

Assignees

• 成都铭鑫时代智慧科技有限公司

Dates

Publication Date

20260512

Application Date

20260204

Claims (10)

1. 1. An intelligent control system for a liquid crystal panel conveying path based on visual identification is characterized by comprising the following steps: S1, recognizing an image of a liquid crystal panel to obtain panel space state information; S2, judging the space state according to the panel space state information to obtain visual constraint information; s3, carrying out path matching analysis according to the visual constraint information to obtain a path adaptation result; s4, judging the path feasibility according to the path adaptation result to obtain a candidate conveying path set; s5, carrying out path cost analysis according to the candidate conveying path set to obtain a target conveying path; s6, generating control instructions according to the target conveying path to obtain a path control result.
2. 2. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 1, wherein the image recognition in the step S1 comprises: And selecting the conveying node as a vision acquisition node, acquiring a node image containing the whole outline of the liquid crystal panel when the liquid crystal panel passes through the node, performing image analysis processing, identifying an edge outline image of the liquid crystal panel, and mapping image coordinates to a physical coordinate system of a conveying device to obtain panel space state information at the conveying node.
3. 3. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 1, wherein the determining of the spatial state in step S2 comprises: And judging the position, the posture and the form of the liquid crystal panel at the conveying node according to the obtained panel space state information, and clearly distinguishing the normal state from the abnormal state to form visual constraint information for conveying path control.
4. 4. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 1, wherein the path matching analysis in step S3 comprises: And carrying out matching analysis on the structural characteristics of the candidate paths by a step-by-step matching method of constraint and path structures according to the obtained visual constraint information, and definitely and structurally fitting the paths of the current liquid crystal panel to obtain a path adaptation result.
5. 5. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 4, wherein the step-by-step matching method of constraint and path structure comprises: Based on position constraint in visual constraint information, matching the bearing range of each conveying path, excluding paths which do not meet the current position requirement of the panel, based on the basis, matching the steering structure of the rest paths based on gesture constraint, excluding paths which do not meet the gesture requirement of the panel, further based on morphological constraint, matching the supporting structure of the rest paths, and summarizing to obtain a path adaptation result.
6. 6. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 1, wherein the path feasibility determining in step S4 comprises: and according to the obtained path adaptation result, combining the conveying state information of the conveying paths, and carrying out feasibility judgment on the paths based on the occupied state and the running stable state by a step-by-step judgment method of multi-state conditions to obtain a candidate conveying path set.
7. 7. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 1, wherein the path cost analysis in step S5 comprises: and calculating path cost values of all paths according to the obtained candidate conveying path set, comparing the path cost values, and realizing path cost analysis to obtain a target conveying path which is most suitable for conveying the current liquid crystal panel.
8. 8. The intelligent control system for a liquid crystal panel conveying path based on visual recognition according to claim 7, wherein the path cost value calculation comprises: ; Wherein: Conveying path Path cost value of (2); Conveying path Is not limited, the conveying time of the conveyor belt is not limited; visual constraints affect the intensity; Time cost weight; Conveying path Is a structural sensitivity coefficient of (2); and comprehensively calculating the path cost value of the conveying path by combining the conveying time, the path structure and the visual constraint of the conveying path.
9. 9. The intelligent control system for a conveying path of a liquid crystal panel based on visual recognition according to claim 1, wherein the generation of the control command in the step S6 comprises: And generating a conveying control instruction sequence corresponding to the path according to the obtained target conveying path, and driving conveying equipment to complete the path guidance of the liquid crystal panel to obtain a path control result.
10. 10. The intelligent control system of a liquid crystal panel conveying path based on visual recognition according to claim 1, comprising the following functional modules: the system comprises a panel image recognition module, a visual constraint judging module, a path structure matching module, a path feasibility judging module, a path cost calculating module and a conveying control executing module.

Description

Liquid crystal display panel conveying path intelligent control system based on visual recognition Technical Field The invention relates to the technical field of visual identification, in particular to an intelligent control system for a liquid crystal panel conveying path based on visual identification. Background The liquid crystal panel is often required to be frequently transported between a plurality of processes in the manufacturing process. Because the liquid crystal panel has the structural characteristics of large size, thin thickness, limited rigidity and the like, the liquid crystal panel has higher requirements on the stability of the posture and the integrity of the shape in the conveying process. Once attitude deflection or local morphological abnormality occurs in the conveying process, stress concentration is easily generated in steering, supporting or clamping links, so that panel damage is caused, and the product yield is affected. Along with the development of computer vision technology, part of conveying systems start to introduce vision recognition means, detect the position or the appearance of a liquid crystal panel, and accordingly perform conveying control. In the prior art, the conveying path matching analysis has the defects that the conventional liquid crystal panel conveying system mostly adopts a preset conveying path or a path control mode based on fixed rules, namely, a plurality of conveying paths are planned in advance in the equipment design stage, one of the conveying paths is selected for conveying according to the process requirements in the operation process, a mechanism for matching and analyzing the visual constraint information and the structural characteristics of the conveying paths is lacking, and the structural adaptation degree of different conveying paths to the current panel state cannot be effectively evaluated, so that the applicability and the conveying stability of the path selection are influenced. Disclosure of Invention Aiming at the defects of the prior art, the invention provides an intelligent control system for a liquid crystal panel conveying path based on visual identification, which aims to solve the problems in the prior art. In order to achieve the above purpose, the present invention provides the following technical solutions: in a first aspect, the present invention provides an intelligent control system for a conveying path of a liquid crystal panel based on visual recognition, including the following steps: S1, recognizing an image of a liquid crystal panel to obtain panel space state information; S2, judging the space state according to the panel space state information to obtain visual constraint information; s3, carrying out path matching analysis according to the visual constraint information to obtain a path adaptation result; s4, judging the path feasibility according to the path adaptation result to obtain a candidate conveying path set; s5, carrying out path cost analysis according to the candidate conveying path set to obtain a target conveying path; s6, generating control instructions according to the target conveying path to obtain a path control result. Further optimizing the technical scheme, the image recognition in the step S1 includes: And selecting the conveying node as a vision acquisition node, acquiring a node image containing the whole outline of the liquid crystal panel when the liquid crystal panel passes through the node, performing image analysis processing, identifying an edge outline image of the liquid crystal panel, and mapping image coordinates to a physical coordinate system of a conveying device to obtain panel space state information at the conveying node. Further optimizing the technical scheme, the space state determination in the step S2 includes: And judging the position, the posture and the form of the liquid crystal panel at the conveying node according to the obtained panel space state information, and clearly distinguishing the normal state from the abnormal state to form visual constraint information for conveying path control. Further optimizing the technical scheme, the path matching analysis in the step S3 includes: And carrying out matching analysis on the structural characteristics of the candidate paths by a step-by-step matching method of constraint and path structures according to the obtained visual constraint information, and definitely and structurally fitting the paths of the current liquid crystal panel to obtain a path adaptation result. Further optimizing the technical scheme, the step-by-step matching method of the constraint and path structure comprises the following steps: Based on position constraint in visual constraint information, matching the bearing range of each conveying path, excluding paths which do not meet the current position requirement of the panel, based on the basis, matching the steering structure of the rest paths based on gesture constraint, excluding paths wh