CN-121994168-A - Confocal microscope-based knife edge diaphragm area testing device and method

CN121994168ACN 121994168 ACN121994168 ACN 121994168ACN-121994168-A

Abstract

The invention provides a confocal microscope-based knife edge diaphragm area testing device and method, and belongs to the technical field of metering testing. During measurement, the reticle surface of the reticle is attached to the measured surface of the knife edge diaphragm, a plurality of partial images with overlapping areas are collected along the edge of the diaphragm through a confocal microscope, the images are accurately spliced into a complete edge image through an automatic splicing algorithm based on sliding window secondary optimization, pixel scale calibration is conducted through the reticle distance from the source tracing to the geometric standard on the reticle, and finally the accurate area is obtained through counting the number of pixels in the light passing area. The method and the device effectively avoid approximation errors and accumulated errors of the traditional diameter integration method by directly measuring all edges and tracing to the length standard, and realize high-precision and traceable measurement of the area of the knife edge diaphragm.

Inventors

GAO YIFAN
WAN YU
CAI JING
ZHANG YANG
WEN YUE
ZHANG XUECONG

Assignees

中国航空工业集团公司北京长城计量测试技术研究所

Dates

Publication Date: 20260508
Application Date: 20251208

Claims (10)

1. A confocal microscope-based knife-edge diaphragm area test method, which is characterized by comprising the following steps: step S1, attaching a measured surface of a knife edge diaphragm to a scribing surface of a reticle, and placing the surface on a stage of a confocal microscope to adjust to a confocal state; S2, continuously collecting a plurality of partial overlapped images along the edge of the knife-edge diaphragm; step S3, automatically splicing all the collected partial images to generate a complete knife-edge diaphragm edge image; S4, extracting a reticle of the reticle in the edge image of the knife edge diaphragm, and calibrating pixel equivalent of the image based on actual interval of the reticle; s5, recognizing and extracting the outline of the light passing area of the knife-edge diaphragm in the edge image of the knife-edge diaphragm, and further counting the number of pixel points in the light passing area; And S6, calculating the actual light passing area of the knife-edge diaphragm based on the pixel number and the pixel equivalent.
2. The method according to claim 1, characterized in that in step S3, all the acquired partial images are automatically stitched, in particular comprising: s31, creating and initializing a variable figureNew for storing the final spliced image; S32, reading a first original image and copying the first original image into figureNew; S33, reading the next original image, and calculating offset dx of two adjacent images in the x direction and offset dy of the two adjacent images in the y direction; s34, expanding figureNew the size according to the x-direction offset dx and the y-direction offset dy, and superposing the current image on the corresponding position of figureNew; and S35, repeatedly executing the steps S33 and S34 until all the original images are spliced.
3. The method according to claim 2, wherein in step S33, the x-direction offset dx and the y-direction offset dy are calculated, comprising the steps of: Performing row and column dimension reduction on two adjacent images f1 and f2 respectively to obtain dimension reduction data f1x and f2x in the x direction and dimension reduction data f1y and f2y in the y direction; and adopting a sliding window method to respectively perform twice optimization calculation on the dimensionality reduction data of the x and y directions so as to determine the offset dx and the offset dy of the adjacent two images in the x direction.
4. A method according to claim 3, characterized in that the two optimization calculations are in particular: Taking the ratio of the average value of the absolute values of the dimension-reduction differences in a certain direction in the sliding window to the average value of the sum as an objective function, and searching the minimum value of the average value to determine the rough offset between the images; and in a limited area containing the rough offset, searching the minimum value of the standard deviation of the dimension reduction data difference in the sliding window to determine the offset of two adjacent images in the direction by taking the standard deviation of the dimension reduction data difference in the sliding window as an objective function.
5. The method according to claim 1, wherein in the step S2, the overlapping area between adjacent partial images is not less than 40% of the area of a single image.
6. The method according to claim 1, wherein in the step S5, a Canny edge detection algorithm is used to extract the profile of the light passing area of the knife-edge diaphragm.
7. The method according to claim 1, wherein in step S5, after extracting the outline of the light-passing area, the outline is further screened by geometric features of the outline, wherein the geometric features include the closure, the area and the circularity of the outline.
8. A confocal microscope-based knife-edge diaphragm area testing device, characterized in that the testing device comprises: The first processing module is configured to attach the measured surface of the knife edge diaphragm to the scribing surface of the reticle, and is arranged on the objective table of the confocal microscope to adjust to a confocal state; a second processing module configured to continuously acquire a plurality of partially overlapping partial images along the edge of the knife-edge diaphragm; The third processing module is configured to automatically splice all the acquired partial images to generate a complete knife-edge diaphragm edge image; a fourth processing module configured to extract a reticle of the reticle in the edge image of the knife-edge diaphragm, and to scale pixel equivalents of the image based on actual spacing of the reticle; The fifth processing module is configured to identify and extract the outline of the light passing area of the knife-edge diaphragm in the edge image of the knife-edge diaphragm, and further count the number of pixel points in the light passing area; and the sixth processing module is configured to calculate the actual light passing area of the knife-edge diaphragm based on the pixel number and the pixel equivalent.
9. An electronic device comprising a memory and a processor, the memory storing a computer program, the processor implementing the steps of a confocal microscope-based knife-edge aperture area test method of any one of claims 1 to 7 when the computer program is executed.
10.A computer readable storage medium, characterized in that it has stored thereon a computer program which, when executed by a processor, implements the steps of a confocal microscope based knife-edge aperture area test method according to any one of claims 1 to 7.

Description

Confocal microscope-based knife edge diaphragm area testing device and method Technical Field The invention belongs to the technical field of metering test, and particularly relates to a confocal microscope-based knife-edge diaphragm area test device and method. Background The traditional area measurement adopts measurement diameter and roundness to calculate the area. Taking a 5mm diaphragm as an example, the uncertainty of the required area is 3E-4, and the uncertainty of the radius is required to reach 15E-5 and can be ensured after the roundness is converted to 0.75 mu m. Conventional machining means have difficulty in ensuring that the roundness of the knife edge reaches this level. The current common area measurement method comprises the steps of measuring the effective area of a diaphragm based on a laser scanning method, taking focused laser as a light source to measure the area, taking an optical fiber focused halogen tungsten lamp as the light source to measure the area, and the like. The method comprises the steps of measuring the effective area of a diaphragm based on a laser scanning method, calculating the area by using energy integration, ensuring the geometric quantity by a high-precision positioning table, and measuring accumulated errors when a large area is measured, wherein the measuring area of a focused laser and an optical fiber focused halogen tungsten lamp is solved by measuring a plurality of diameter integrals, so that the influence of the edge of the diaphragm is ignored. Disclosure of Invention Aiming at the technical problem of inaccurate measurement of the area of a knife-edge diaphragm, the invention provides a device and a method for measuring the area of the knife-edge diaphragm based on a confocal microscope. The invention discloses a confocal microscope-based knife-edge diaphragm area testing method, which comprises the following steps: step S1, attaching a measured surface of a knife edge diaphragm to a scribing surface of a reticle, and placing the surface on a stage of a confocal microscope to adjust to a confocal state; S2, continuously collecting a plurality of partial overlapped images along the edge of the knife-edge diaphragm; step S3, automatically splicing all the collected partial images to generate a complete knife-edge diaphragm edge image; S4, extracting a reticle of the reticle in the edge image of the knife edge diaphragm, and calibrating pixel equivalent of the image based on actual interval of the reticle; s5, recognizing and extracting the outline of the light passing area of the knife-edge diaphragm in the edge image of the knife-edge diaphragm, and further counting the number of pixel points in the light passing area; And S6, calculating the actual light passing area of the knife-edge diaphragm based on the pixel number and the pixel equivalent. Optionally, in step S3, all the collected partial images are automatically stitched, which specifically includes: s31, creating and initializing a variable figureNew for storing the final spliced image; S32, reading a first original image and copying the first original image into figureNew; S33, reading the next original image, and calculating offset dx of two adjacent images in the x direction and offset dy of the two adjacent images in the y direction; s34, expanding figureNew the size according to the x-direction offset dx and the y-direction offset dy, and superposing the current image on the corresponding position of figureNew; and S35, repeatedly executing the steps S33 and S34 until all the original images are spliced. Optionally, in step S33, the x-direction offset dx and the y-direction offset dy are calculated, including the steps of: Performing row and column dimension reduction on two adjacent images f1 and f2 respectively to obtain dimension reduction data f1x and f2x in the x direction and dimension reduction data f1y and f2y in the y direction; and adopting a sliding window method to respectively perform twice optimization calculation on the dimensionality reduction data of the x and y directions so as to determine the offset dx and the offset dy of the adjacent two images in the x direction. Optionally, the two optimizing calculations are specifically: Taking the ratio of the average value of the absolute values of the dimension-reduction differences in a certain direction in the sliding window to the average value of the sum as an objective function, and searching the minimum value of the average value to determine the rough offset between the images; and in a limited area containing the rough offset, searching the minimum value of the standard deviation of the dimension reduction data difference in the sliding window to determine the offset of two adjacent images in the direction by taking the standard deviation of the dimension reduction data difference in the sliding window as an objective function. Optionally, in the step S2, an overlapping area between adjacent partial images is not less