Automatic fence correction method based on image processing
Through the combination of YOLOv8 image processing algorithm and robotic arm, the automatic high-precision correction of microarray chip fences is achieved, solving the problems of manual alignment error and insufficient automated detection, and improving operational efficiency and alignment effect.
Patent Information
- Application Number
- CN202510299673.X
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-03-14
- Publication Date
- 2025-07-08
AI Technical Summary
In the prior art, the fence placement of the microarray chip has problems such as large manual alignment error, insufficient automated detection accuracy and inability to correct automatically, resulting in low operating efficiency and poor alignment effect.
The YOLOv8 image processing algorithm is used to detect the inclination angle of the fence, and combined with the automated robot arm for autonomous correction, so as to achieve accurate alignment of the fence through image acquisition, object detection, angle calculation and distance calculation.
Fully automatic and high-precision fence correction is achieved, which improves operating efficiency and ensures accurate alignment of fences with solid-phase carriers without manual intervention.
Smart Images
Figure CN120279237A_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of biochip manufacturing, and specifically to an automatic fence correction method based on image processing. Background Art
[0002] According to different structures, common biochips include microfluidic chips, microarray chips, etc. Among them, a microarray chip is to fix biomolecules (DNA, protein, cells, etc.) on the surface of a solid-phase carrier (such as a glass slide, silicon wafer, nylon membrane) in a two-dimensional array form (flow channel). When in use, it generally matches a fence to form a complete microarray chip product. Multiple square sample holes (or some are circular sample holes) are cut on the fence. When in use, various samples to be detected can be added into the sample holes respectively, and after reacting with the biomolecules on the solid-phase carrier, observation and analysis can be carried out.
[0003] When manufacturing a microarray chip product, it is necessary to strictly align the fence with the solid-phase carrier to ensure that the biomolecules on the solid-phase carrier correspond to the sample holes of the fence, so as to ensure effective and accurate results when the chip is used. Currently, in the existing fence placement scenarios, it usually relies on manual measurement and alignment of the solid-phase carrier, resulting in high errors and low efficiency. Some existing automated devices can detect the fence through simple image recognition, but cannot achieve high-precision tilt correction. The specific technical problems include: 1. Manual alignment error: Traditional manual alignment methods require operators to manually measure and adjust, which is prone to errors, especially in large-scale placement scenarios, which are very time-consuming and laborious; 2. Insufficient accuracy of automated detection: Current automatic detection technologies can often only identify the position of the fence, and it is difficult to accurately determine the angle and tilt deviation of the fence, resulting in poor alignment effects; 3. Unable to correct autonomously: Existing technologies cannot perform real-time autonomous adjustment in combination with the detected angle deviation, and manual intervention is required. The cleanliness requirements for fence preparation are high, and it is rather cumbersome for personnel to enter and exit the workshop. Summary of the Invention
[0004] In order to solve the above problems, the present invention provides an automatic fence correction method based on image processing, which detects the fence tilt angle through the YOLOv8 image processing algorithm and performs autonomous correction in combination with an automated robotic arm. The technical solution adopted by the present invention is as follows: An automatic fence correction method based on image processing, comprising the following steps: S1) Image acquisition, acquiring real-time images of the fence and the fixed area; S2) Target detection, identifying and annotating the sample holes of the fence, and at the same time identifying the fixed area in the image; S3) Distance calculation: After the fence attitude correction is completed, calculate the relative distances between the fence and the fixed area along the X-axis and Y-axis. and S4) Robot arm correction: According to the calculated position deviation, control the robot arm to adjust the fence to the set position within the fixed area.
[0005] Before the above step S3), it also includes S2.5) Angle calculation: By detecting the boundary coordinates of the sample holes of the fence, calculate its tilt angle; in the above step S4), the robot arm first performs angle correction according to the calculated tilt angle and then performs position adjustment.
[0006] The process of the above step S2.5) Angle calculation is as follows: Detect two coordinates (x1, y1) and (x4, y4) of the boundary of the sample hole, and use the formula to calculate the tilt angle θ of the fence relative to the horizontal line. The process of the above step S3) Distance calculation is as follows: S3.1) First calculate the center point coordinates. Assume the center point coordinates of the fence are (Xw, Yw), and calculate the center of the fence through the detected boundary of the sample hole; the center point coordinates of the fixed area are (Xf, Yf); S3.2) Then calculate the relative distance. The offset in the X-axis direction is ∆X = Xf - Xw, and the offset in the Y-axis direction is ∆Y = Yf - Yw. Based on ∆X and ∆Y, calculate the relative distances between the fence and the fixed area in the X-axis and Y-axis directions. Based on the calculations of step S2.5) and step S3), in step S4) Robot arm correction, first adjust the fence attitude until the tilt angle θ = 0, at which time the fence is parallel to the horizontal plane; then move the fence to the set position within the fixed area according to ∆X and ∆Y to ensure that the fence is aligned with the target area.
[0007] The beneficial effects of the present invention are as follows: First, full-automatic correction: By accurately detecting the sample holes and ensuring accurate correction of the tilt angle and position of the fence, it can automatically complete the tilt angle correction and position alignment without manual intervention, greatly improving the operation efficiency.
[0008] Second, dual correction function: It can not only calculate the relative distance from the fixed area to ensure that the fence is adjusted to the predetermined position to accurately align with the target area, but also correct the attitude of the fence, which is applicable to high-precision fence alignment scenarios. Brief Description of the Drawings
[0009] Figure 1 is the image of the fence position before automatic correction of the present invention; Figure 2 is the image of the fence position after automatic correction of the present invention. Detailed Embodiments
[0010] The present invention will be described in detail below with reference to the accompanying drawings. The following embodiments are all illustrative and are intended to provide further description of the present application. Unless otherwise specified, all technical terms used have the same meaning as commonly understood by those skilled in the technical field to which the present application belongs. It should be noted that the terms used are only for describing specific embodiments and are not intended to limit the present application.
[0011] This embodiment is an automatic fence correction method based on image processing, including the following steps: S1) Image acquisition, the camera carried by the robotic arm acquires real-time images of the fence surface and the fixed area. S2) Target detection, the YOLOv8 algorithm is used to analyze the image to identify the sample holes and the fixed area of the fence, and the boundaries of the sample holes are marked (similar Figure 2 state), and the fence in this embodiment has square sample holes. S2.5) Angle calculation, by detecting the boundary coordinates of the sample holes of the fence, calculate its tilt angle. S3) Distance calculation, after the fence attitude correction is completed, calculate the relative distances of the X-axis and Y-axis between the fence and the fixed area. and S4) Robotic arm correction, the robotic arm first performs angle correction according to the calculated tilt angle, and then controls the robotic arm to adjust the fence to the set position within the fixed area according to the calculated position deviation.
[0012] Specifically, the process of the above step S2.5) angle calculation is as follows: Detect two coordinates (x1, y1) and (x4, y4) of the boundary of the square sample hole, for example, they can be two vertices of the sample hole, and use the formula to calculate the tilt angle θ of the fence relative to the horizontal line.
[0013] The process of the above step S3) distance calculation is as follows: S3.1) First calculate the center point coordinates. Assume that the center point coordinates of the fence are (Xw, Yw), and calculate the center of the fence through the detected boundary of the square sample hole; the center point coordinates of the fixed area are (Xf, Yf); S3.2) Then calculate the relative distance. The offset in the X-axis direction is ∆X = Xf - Xw, and the offset in the Y-axis direction is ∆Y = Yf - Yw. Based on ∆X and ∆Y, calculate the relative distances of the fence and the fixed area in the X-axis and Y-axis directions.
[0014] Based on the calculations of step S2.5) and step S3), in step S4) robotic arm correction, first adjust the fence attitude until the tilt angle θ = 0, at this time the fence is parallel to the horizontal plane; then move the fence to the set position within the fixed area according to ∆X and ∆Y to ensure that the fence is aligned with the target area.
[0015] This automatic correction method can not only detect the tilt angle of the fence and perform automatic correction, but also calculate the relative distance between the corrected fence and the fixed area, control the robotic arm to move the fence to a predetermined position, without manual operation, greatly improving efficiency, and is applicable to the high-precision correction and alignment of the fence.
[0016] The above are only the preferred embodiments of the present application. It should be noted that for those of ordinary skill in the art, without departing from the principle of the present application, several changes or improvements can be made, and these changes or improvements should also be regarded as the protection scope of the present application.
Claims
1. An automatic fence correction method based on image processing, characterized in that Including the following steps: S1) Image acquisition, acquiring real-time images of the fence and the fixed area; S2) Target detection, identifying and labeling the sample holes of the fence, and at the same time identifying the fixed area in the image; S3) Distance calculation, after the fence attitude correction is completed, calculating the relative distances between the fence and the fixed area in the X-axis and Y-axis directions; and S4) Robot arm correction, according to the calculated position deviation, controlling the robot arm to adjust the fence to the set position within the fixed area.
2. The fence automatic correction method based on image processing according to claim 1, characterized in that, Before the above step S3), it also includes S2.5) Angle calculation, by detecting the boundary coordinates of the sample holes of the fence, calculating its tilt angle; In the above step S4), the robot arm first performs angle correction according to the calculated tilt angle and then performs position adjustment.
3. The fence automatic correction method based on image processing according to claim 2, wherein The process of the above step S2.5) for angle calculation is as follows: Detect two coordinates (x1, y1) and (x4, y4) of the boundary of the sample hole, and use the formula to calculate the inclination angle θ of the fence relative to the horizontal line; The process of the above step S3) distance calculation is as follows: S3.1) First calculate the center point coordinates. Assuming the center point coordinates of the fence are (Xw, Yw), calculate the center of the fence through the detected sample hole boundaries; the center point coordinates of the fixed area are (Xf, Yf); S3.2) Then calculate the relative distances. The offset in the X-axis direction is ∆X = Xf - Xw, and the offset in the Y-axis direction is ∆Y = Yf - Yw. Based on ∆X and ∆Y, calculate the relative distances between the fence and the fixed area in the X-axis and Y-axis directions; Based on the calculations of step S2.5) and step S3), in step S4) robot arm correction, first adjust the fence attitude until the tilt angle θ = 0, at this time the fence is parallel to the horizontal plane; then move the fence to the set position within the fixed area according to ∆X and ∆Y to ensure that the fence is aligned with the target area.