CN-122023749-A - Visual positioning method and system based on ROI dynamic alignment

CN122023749ACN 122023749 ACN122023749 ACN 122023749ACN-122023749-A

Abstract

The invention discloses a visual positioning method and a visual positioning system based on dynamic alignment of ROI (region of interest), and relates to the technical field of machine vision. The method comprises the steps of moving a camera to a predicted position of a marking point, calculating offset between the marking point and the center of a visual field, driving the camera to move the searching point according to a specific path (such as a 'back' path) in a preset local area if the offset is too large, moving the marking point to the center of the visual field through mechanical compensation after the marking point is found, and finally carrying out high-precision identification in a low-distortion area of the center. The invention dynamically adjusts the ROI to ensure that the mark point is always identified in the optimal imaging area, thereby fundamentally eliminating the edge effect caused by the edge distortion of the lens, realizing high-precision, high-efficiency and low-cost visual positioning under the condition of not needing a telecentric lens or a complex distortion model, and being particularly suitable for alignment procedures in the high-end precision manufacturing fields of semiconductors, display panels and the like.

Inventors

YANG YANG
XU HUANSU
CHEN SHUBIN

Assignees

东莞触点智能装备有限公司

Dates

Publication Date: 20260512
Application Date: 20260203

Claims (8)

1. The visual positioning method based on the dynamic alignment of the ROI is applied to a visual positioning system comprising a bearing platform, a visual imaging unit, a driving mechanism and a control unit, and a workpiece to be aligned is positioned through the bearing platform, and is characterized by comprising the following steps of: s1, a control unit controls the driving mechanism to operate a bearing platform with a workpiece positioned to a preset starting position according to original preset parameters, wherein the starting position is positioned in the visual field range of a visual imaging unit; S2, a camera in the visual imaging unit shoots images of the workpieces in the visual field, and whether marking points on the workpieces exist in the range of an image observation area shot by the camera is judged; If the judgment result is no, the step S3 is entered, If yes, entering step S4; S3, the control unit controls the driving mechanism, and path traversing movement is carried out in the local area around the initial position by taking the initial position as a starting point, and images are continuously shot through the camera in the moving process so as to find marking points; If a marker point is found in the local area, the process proceeds to step S4, If no mark point is found in the local area, judging that the positioning fails at the present time, and resetting a preset starting position; s4, determining a coordinate difference value between the center of the mark point and the center of the visual field shot by the camera, and judging whether the difference value is smaller than a set threshold value or not; if the difference is greater than or equal to the set threshold, step S5 is entered, If the difference value is smaller than the set threshold value, the step S6 is entered; s5, if the difference value is larger than or equal to the set threshold value, controlling the driving mechanism to mechanically compensate the difference value through the control unit so that the difference value is smaller than the set threshold value; and S6, if the difference value is smaller than the set threshold value, positioning is completed.
2. The method of claim 1, wherein in step S3, the path traversing means adopts a "loop" path.
3. The method for dynamically aligning visual positioning based on ROI according to claim 1, wherein in the step S3, the preset local area is a rectangular area with a starting point position as a center and a side length of 3-10 mm.
4. A visual positioning method based on dynamic ROI alignment according to claim 3, wherein the preset local area is a square area of 5mm by 5mm.
5. The method for dynamically aligning a visual positioning system according to claim 1 wherein the image observation area is a low distortion area located at the center of the camera field of view.
6. The method of claim 1, wherein after the step S5 is completed, the step S4 is returned again, the coordinate difference between the center of the marker point and the center of the field of view photographed by the camera is recalculated, and the judgment is performed again.
7. The method of claim 1, wherein the visual imaging unit comprises at least one of a camera above the carrying platform and a camera below the carrying platform.
8. A system for a visual localization method based on dynamic alignment of ROIs, the system comprising: the bearing platform is used for bearing and positioning workpieces to be aligned; the visual imaging unit comprises at least one of a camera positioned above the bearing platform and a camera positioned below the bearing platform, and the added light paths point to the bearing platform; The driving mechanism is connected with the bearing platform and used for driving the bearing platform to move in a plane; The control unit is electrically connected with the driving mechanism; The system employs a visual positioning method based on ROI dynamic alignment as claimed in any one of claims 1 to 6.

Description

Visual positioning method and system based on ROI dynamic alignment Technical Field The invention relates to the technical field of machine vision and precise automation equipment, in particular to a visual positioning method and system based on dynamic alignment of ROI. Background In the field of manufacturing high-end equipment such as semiconductor packaging, high-precision mounting and the like, visual alignment is a key step for realizing precision machining and assembly. For example, the current mainstream method of semiconductor lamination alignment is to take a Mark (i.e. a Mark point on a wafer or a chip) on a workpiece by a fixed camera, calculate its central coordinate by using a Pattern Recognition (PR) algorithm, and compare it with a target position, so as to guide a mechanical platform to perform compensation alignment. The method has the inherent problem of 'edge effect', namely, because of image distortion commonly existing in an optical lens, when a mark point is positioned in the edge area of the field of view of a camera, imaging of the mark point is distorted, so that a larger error is generated between a central coordinate calculated by a PR algorithm and an actual coordinate, and the overall alignment precision of equipment is severely restricted. To circumvent the above problems, two schemes are generally adopted in the prior art: The scheme one adopts the telecentric lens with high price to reduce the image distortion, but the cost of the system is greatly increased; in the second scheme, a complex full-field distortion correction model is established for real-time correction, but the calculation complexity and the calibration time of the system are obviously increased, and the production efficiency is reduced. Therefore, a new visual positioning scheme which can ensure high alignment precision and has the advantages of high efficiency and low cost is needed in the field. Disclosure of Invention The invention aims to solve the technical problem of overcoming the defects of the prior art and providing a visual positioning method and a visual positioning system based on dynamic alignment of ROI. In order to solve the technical problems, the positioning method adopts the following technical scheme that the visual positioning method based on the dynamic alignment of the ROI is applied to a visual positioning system comprising a bearing platform, a visual imaging unit, a driving mechanism and a control unit, and a workpiece to be aligned is positioned through the bearing platform; S2, a camera in a visual imaging unit shoots images of a workpiece in a visual field, judging whether a mark point on the workpiece exists in an image observation area range shot by the camera, if the judgment result is NO, entering a step S3, if the judgment result is NO, entering a step S4, S3, a control unit controls a driving mechanism to take the initial position as a starting point, carry out path traversing movement around the initial position in a local area, continuously shoot images through the camera in the moving process so as to search the mark point, if the mark point is found in the local area, entering the step S4, if the mark point is not found in the local area, judging that the positioning is failed at the present time, resetting the preset initial position, S4, determining a coordinate difference value between the center of the mark point and the center of the visual field shot by the camera, judging whether the difference value is smaller than a set threshold, entering the step S5, if the difference value is smaller than the set threshold, entering the step S6, if the difference value is larger than or equal to the set threshold, carrying out mechanical compensation on the difference value through the control unit, and S6, if the difference value is smaller than the set threshold value, completing positioning. Further, in the above technical solution, in step S3, the path traversing movement adopts a "loop" path. In the above technical solution, in step S3, the preset local area is a rectangular area with a starting point position as a center and a side length of 3-10 mm. Further, in the above technical solution, the preset local area is a square area of 5mm×5 mm. Further, in the above technical solution, the image observation area refers to a central low distortion area located in the field of view of the camera. In the above technical solution, after the step S5 is completed, the process returns to the step S4 again, and the coordinate difference between the center of the mark point and the center of the field of view captured by the camera is recalculated, and then the judgment is performed again. Further, in the technical scheme, the visual imaging unit comprises at least one of a camera positioned above the bearing platform and a camera positioned below the bearing platform. The positioning system adopts the following technical scheme that the system comprises a bearing platform, a vis