Intelligent optical microscope automatic alignment system

Abstract

The invention relates to an intelligent optical microscope automatic alignment system, which comprises an image preprocessing module, a straight line detection and feature extraction module, a bias amount calculation module and a circulation adjustment module, the linear detection and feature extraction module automatically detects the optimal threshold value of an image parameter and performs feature point extraction, the bias amount calculation module calculates the bias amount of the image, the bias amount is circularly and continuously calculated for multiple times through the circular adjustment module, a dynamic adjustment strategy is adopted for optimization, a stop condition is achieved, and an optimal alignment position is found. According to the automatic alignment system of the intelligent optical microscope, automatic alignment of the optical microscope is achieved through the artificial intelligence technology, and manual operation is reduced; the alignment precision is improved through image processing and straight line detection, and errors are reduced; the alignment operation is simplified, the technical threshold is reduced, and the user experience is improved.

Description

Technical Field

[0001] The invention relates to the field of optical microscopes, and in particular to an intelligent optical microscope automatic alignment system. Background Art

[0002] The traditional optical microscope alignment process of the wafer is cumbersome and relies on manual operation, which requires the following steps: Figure 1 As shown, at a small magnification, the stage moves to the right side of the image, and the alignment point is manually found and clicked to center. At a small magnification, the stage moves to the right side of the image, and the alignment point is manually found and clicked to center. Figure 2 As shown in the figure, at high magnification, the alignment point is manually found on the right side of the image and the center is clicked; at high magnification, the stage returns to the left side of the image, and the alignment point is manually found and the center is clicked. This process requires multiple steps of manual operation, which is time-consuming and inefficient; and manual alignment is prone to errors, affecting the accuracy of the observation results; multiple steps will also increase the complexity of the operation and the training cost. Summary of the invention

[0003] In order to solve the technical problems existing in the prior art, the present invention provides an intelligent optical microscope automatic alignment system.

[0004] To achieve the above object, the technical solution of the present invention is as follows:

[0005] An intelligent optical microscope automatic alignment system comprises an image preprocessing module, a straight line detection and feature extraction module, a bias calculation module and a loop adjustment module. The image preprocessing module performs contrast enhancement and denoising on an optical microscope image. The straight line detection and feature extraction module automatically detects an optimal threshold of image parameters and extracts feature points. The bias calculation module calculates the image bias. The loop adjustment module continuously calculates the bias through multiple cycles and adopts a dynamic adjustment strategy to optimize, so as to reach a stop condition and find the best alignment position.

[0006] As a preferred technical solution, the image preprocessing module enhances the image contrast by using histogram equalization and gamma correction to improve the clarity of edge detection.

[0007] As a preferred technical solution, the image preprocessing module introduces Gaussian blur or median filtering to the image denoising process to reduce the impact of image noise on edge detection.

[0008] As a preferred technical solution, in the line detection and feature extraction module, line detection includes combining adaptive Hough transform, optimizing parameter selection, and automatically detecting the best threshold.

[0009] As a preferred technical solution, in the line detection and feature extraction module, feature point extraction adopts SIFT or ORB feature detection algorithm.

[0010] As a preferred technical solution, the formula for calculating the bias amount by the bias amount calculation module is:

[0011] x = horizontal line position - ideal horizontal line position x = \text{horizontal line position} - \text{ideal horizontal line position} x = horizontal line position - ideal horizontal line position;

[0012] y = vertical line position - ideal vertical line position y = \text{vertical line position} - \text{ideal vertical line position} y = vertical line position - ideal vertical line position;

[0013] r=|detected angle - ideal angle|r=|\text{detected angle} -\text{ideal angle}|r=|detected angle - ideal angle|;

[0014] Where x is the horizontal distance between the detected horizontal line and the ideal horizontal line, y is the vertical distance between the detected vertical line and the ideal vertical line; and r is the difference between the detected tilt angle and the ideal angle.

[0015] As a preferred technical solution, the cyclic adjustment module reaches a stop condition, which means that when the values ​​of x, y, and r are all close to 0, the adjustment is stopped and the optimal alignment position is determined.

[0016] As a preferred technical solution, the dynamic adjustment strategy includes genetic algorithm or particle swarm optimization.

[0017] Compared with the prior art, the present invention has the following beneficial effects:

[0018] The intelligent optical microscope automatic alignment system of the present invention realizes automatic alignment of the optical microscope through artificial intelligence technology, reducing manual operation; improves alignment accuracy and reduces errors through image processing and straight line detection; simplifies alignment operation, lowers technical threshold, and improves user experience. BRIEF DESCRIPTION OF THE DRAWINGS

[0019] Figure 1 This is an image taken at a small magnification during the alignment process of a wafer using a traditional optical microscope;

[0020] Figure 2 It is an image taken at a high magnification during the alignment process of the wafer using a traditional optical microscope;

[0021] Figure 3 The present invention is a flowchart of the automatic alignment system for an intelligent optical microscope. DETAILED DESCRIPTION

[0022] The technical solution of the present invention is further described below in conjunction with specific implementation methods:

[0023] like Figure 3 As shown in FIG. 1 , an intelligent optical microscope automatic alignment system has a workflow including image preprocessing, line detection and feature extraction, bias calculation and loop adjustment mechanism. Image preprocessing:

[0024] Enhance image contrast: Use histogram equalization and gamma correction to improve the clarity of edge detection.

[0025] Denoising: Introduce Gaussian blur or median filtering to reduce the impact of image noise on edge detection.

[0026] Line detection and feature extraction:

[0027] Improved Hough transform: Combined with adaptive Hough transform, optimized parameter selection, and automatic detection of the best threshold to improve line detection effect.

[0028] Feature point extraction: Use feature detection algorithms such as SIFT or ORB to extract more key feature points to improve positioning accuracy.

[0029] Bias calculation:

[0030] Bias definition:

[0031] xBias: The horizontal distance between the detected horizontal line and the ideal horizontal line.

[0032] y-bias: The vertical distance between the detected vertical line and the ideal vertical line.

[0033] rBias: The difference (error) between the detected tilt angle and the ideal angle.

[0034] Calculation formula:

[0035] x=horizontal line position-ideal horizontal line positionx=\text{horizontal line position}-\text{ideal horizontal line position}x=horizontal line position-ideal horizontal line position

[0036] y = vertical line position - ideal vertical line position y = \text{vertical line position} - \text{ideal vertical line position} y = vertical line position - ideal vertical line position;

[0037] r=|detected angle - ideal angle|r=|\text{detected angle} -\text{ideal angle}|r=|detected angle - ideal angle|

[0038] Cycle adjustment mechanism:

[0039] Feedback loop: Over multiple cycles, the bias (x, y, r) is continuously calculated and the microscope position is fine-tuned.

[0040] Stop condition: When the values ​​of x, y, and r are all close to 0, stop adjusting and determine it as the best alignment position.

[0041] Optimization algorithms: Use dynamic adjustment strategies, such as genetic algorithms or particle swarm optimization, to achieve a more efficient search process.

[0042] Through these optimizations, the system will be able to accurately output the amount of bias and achieve the optimal alignment position over multiple cycles, improving the automation and accuracy of optical microscopy.

[0043] This embodiment is only a further explanation of the invention rather than a limitation of the invention. After reading this specification, those skilled in the art may make non-creative modifications to this embodiment as needed. However, as long as they are within the scope of the claims of the invention, they are protected by the patent law.

Claims

1. An intelligent optical microscope automatic alignment system, characterized in that: The invention comprises an image preprocessing module, a straight line detection and feature extraction module, a bias calculation module and a loop adjustment module. The image preprocessing module performs contrast enhancement and denoising on the optical microscope image. The straight line detection and feature extraction module automatically detects the optimal threshold of the image parameter and extracts feature points. The bias calculation module calculates the image bias. The loop adjustment module continuously calculates the bias through multiple cycles and adopts a dynamic adjustment strategy to optimize, so as to reach the stopping condition and find the optimal alignment position.

2. The intelligent optical microscope automatic alignment system according to claim 1, characterized in that: The image preprocessing module enhances the image contrast by using histogram equalization and gamma correction to improve the clarity of edge detection.

3. The intelligent optical microscope automatic alignment system according to claim 1, characterized in that: The image preprocessing module introduces Gaussian blur or median filtering to the image denoising process, thereby reducing the influence of image noise on edge detection.

4. The intelligent optical microscope automatic alignment system according to claim 1, characterized in that: In the line detection and feature extraction module, line detection includes combining adaptive Hough transform, optimizing parameter selection, and automatically detecting the best threshold.

5. The intelligent optical microscope automatic alignment system according to claim 1, characterized in that: In the line detection and feature extraction module, feature point extraction adopts SIFT or ORB feature detection algorithm.

6. The intelligent optical microscope automatic alignment system according to claim 1, characterized in that: The formula for calculating the bias amount by the bias amount calculation module is: x = horizontal line position - ideal horizontal line position x = \text{horizontal line position} - \text{ideal horizontal line position} x = horizontal line position - ideal horizontal line position; y = vertical line position - ideal vertical line position y = \text{vertical line position} - \text{ideal vertical line position} y = vertical line position - ideal vertical line position; r=|detected angle - ideal angle|r=|\text{detected angle} -\text{ideal angle}|r=|detected angle - ideal angle|; Where x is the horizontal distance between the detected horizontal line and the ideal horizontal line, y is the vertical distance between the detected vertical line and the ideal vertical line; and r is the difference between the detected tilt angle and the ideal angle.

7. The intelligent optical microscope automatic alignment system according to claim 6, characterized in that: The cyclic adjustment module reaches the stop condition when the values ​​of x, y, and r are all close to 0, and stops adjusting to determine the best alignment position.

8. The intelligent optical microscope automatic alignment system according to claim 1, characterized in that: Dynamic adjustment strategies include genetic algorithms or particle swarm optimization.

Images