CN-121994819-A - Surface image acquisition method and surface image acquisition device for conical member

CN121994819ACN 121994819 ACN121994819 ACN 121994819ACN-121994819-A

Abstract

The application discloses a surface image acquisition method and a surface image acquisition device for a conical member, wherein the surface image acquisition method comprises the steps of carrying out line scanning on the conical member which moves rotationally by a line-array camera in an image acquisition unit and a telecentric lens at a fixed line frequency to acquire an original image with regular distortion; establishing a coordinate mapping relation between pixel coordinates of an original image and surface geometric coordinates of the conical member according to the fixed line frequency and the rotary motion of the conical member; and correcting the original image according to the coordinate mapping relation to obtain a target image. The application can obtain the rotation scanning image with constant magnification and no perspective distortion, realizes the pixel level inverse mapping correction based on the cone geometric model, has easy realization, low cost and high reliability, and can realize the sub-millimeter level precision detection.

Inventors

LIU DONG
LI XIYUAN
WANG ZEDONG
WANG SHILING
ZHANG FENGWEI
LIU CHONG

Assignees

浙江大学

Dates

Publication Date: 20260508
Application Date: 20260409

Claims (10)

1. A surface image acquisition method for a tapered member, the surface image acquisition method comprising: the linear array camera in the image acquisition unit is matched with the telecentric lens to perform line scanning on the rotary cone-shaped component at a fixed line frequency, so as to obtain an original image with regular distortion; Establishing a coordinate mapping relation between pixel coordinates of the original image and surface geometric coordinates of the conical member according to the fixed line frequency and the rotary motion of the conical member; And correcting the original image according to the coordinate mapping relation to obtain a target image.
2. The surface image acquisition method according to claim 1, wherein the fixed line frequency is set with reference to a linear velocity of an outer ring edge of the large end of the tapered member at the time of the rotational movement of the tapered member.
3. The surface image acquisition method according to claim 2, wherein the fixed line frequency F satisfies the following formula: ; wherein ω is a rotational angular velocity of the tapered member, L is a bus length of the tapered member, α is a half-apex angle of the tapered member, and σ is an object space resolution of the image capturing unit.
4. The surface image acquisition method according to claim 1, wherein a line camera is set to continuously acquire N lines of images at a fixed line frequency to form the original image, the row index u=0, 1,2, N-1; the column index v=0, 1,2, P-1; wherein P is the resolution of the linear camera; The coordinate mapping relationship is configured to: ; ; Wherein, the For the rotation angle theta, omega corresponding to the image of the u-th line in the original image, the rotation angular velocity of the conical member, The acquisition time from the original image to the image of the ith row is taken as the acquisition time; For the relative position of the projection of the pixel point in the v-th column image in the original image to the bus direction of the conical member, And (3) for the coordinate component of the pixel point in the v-th image in the original image in the axial direction of the conical member, sigma is the object space resolution of the image acquisition unit, L is the bus length of the conical member, and alpha is the half vertex angle of the conical member.
5. The surface image acquisition method according to claim 4, wherein the correcting the original image according to the coordinate mapping relationship includes: Determining a telescoping factor lambda according to the coordinate mapping relation, wherein the telescoping factor lambda is used for representing local geometric telescoping proportion generated by linear speed difference when each pixel row axially distributed along a conical member in the original image is imaged in the circumferential direction; Establishing a corresponding relation between a coordinate system (l, s) of the target image and a coordinate system (theta, s) of the original image by adopting a rectangular unfolding strategy based on arc length-bus coordinates; And according to the corresponding relation and the representation of the pixel coordinates (u, v) of the original image and the pixel coordinates (i, j) of the target image in the respective coordinate system (theta, s) and the coordinate system (l, s), completing the coordinate conversion between the original image and the target image, and generating the target image.
6. The surface image acquisition method according to claim 5, wherein a normalized busbar position S e [ S min , 1] is defined, where s=s min corresponds to the cone apex and s=1 corresponds to the large end, and the expansion factor λ satisfies the following formula: ; Wherein s is the normalized busbar position, Is the expansion factor at position s on the generatrix of the conical member, V(s) is the linear velocity at position s on the generatrix of the conical member, for the telescoping factor at the reference datum, Is the linear velocity of the outer rim edge of the large end of the conical member when the conical member is in rotational motion, Is the normalized position parameter value of the large end of the conical member, ω is the rotational angular velocity of the conical member, L is the generatrix length of the conical member, a is the half apex angle of the conical member, And Are all 1.
7. The surface image acquisition method according to claim 5, wherein the correspondence relationship is: ; where θ is the rotation angle, x(s) is the x-axis coordinate corresponding to a point on the generatrix, and α is the half apex angle of the conical member.
8. The surface image acquisition method as set forth in claim 7, wherein the size of the target image is set to be H x W, H and W are determined according to the pixel accuracy requirement of the target image, H is the number of rows and W is the number of columns; The relation between the pixel coordinates (i, j) of the target image and the coordinate system (l, s) is expressed as: ; ; Wherein i=0, 1,2, H-1;j =0, 1,2, W-1, the scale factor of different positions in each row direction L is the length of a bus of the conical member, and alpha is the half apex angle of the conical member; Arc length range Normalized busbar position 。
9. The surface image acquisition method according to claim 8, wherein the generating the target image from the representation of the pixel coordinates (i, j) of the target image in a coordinate system (i, s) comprises: Traversing pixel coordinates of each target image in a reverse mapping mode, calculating pixel coordinates (i, j) of the target image through an inverse transformation formula, and reversely mapping the pixel coordinates (i, j) of the target image to pixel floating point coordinates of the original image ; The inverse transformation formula can be obtained by inverse solution of the following relation: ; ; inverse mapping to coordinates of the original image Can be expressed as: ; ; based on the principle of regional interpolation, each pixel floating point coordinate Calculating the weighted average of all pixels in an area corresponding to the area with a certain size in the original image to obtain an interpolation pixel value; And giving the interpolation pixel value to a corresponding pixel point (i, j) in the target image to generate the target image.
10. A surface image acquisition device for a conical member, the surface image acquisition device comprising: a rotation driving unit for driving the tapered member to rotate about its own axis; The image acquisition unit comprises a linear array camera and a telecentric lens, wherein the linear array camera is used for carrying out line scanning on a conical component which moves rotationally at a fixed line frequency in cooperation with the telecentric lens to acquire an original image with regular distortion; The data processing unit is used for establishing a coordinate mapping relation between the pixel coordinates of the original image and the pixel coordinates of the surface unfolding image of the conical member according to the fixed line frequency and the rotary motion of the conical member, and correcting the original image according to the coordinate mapping relation and a transformation algorithm to obtain a target image.

Description

Surface image acquisition method and surface image acquisition device for conical member Technical Field The application relates to the technical field of surface defect detection, in particular to a surface image acquisition method and a surface image acquisition device for a conical member. Background Conical or truncated cone-shaped components are commonly used in the fields of aerospace, energy power and the like, such as components of an air inlet pipe, an exhaust pipe, a combustion chamber liner and the like of a turbine engine, and conical or truncated cone-shaped structures are generally adopted for obtaining better rectifying effects. The conical member is generally made of alloy or advanced composite material, and has extremely high requirements on-line reliability, however, various typical defects are easily generated on the surface of the conical member in the processing and manufacturing process due to the special physical, chemical and mechanical properties of the conical member. Such as scratches, cracks, pits, etc. on the metal surface, and bubbles, sloughs, inclusions, wrinkles, scratches, etc. on the carbon fiber composite surface, which may have serious effects in practical applications. It is therefore necessary to perform surface defect detection on the tapered member before shipment. At present, the surface defect of the conical member is generally detected by adopting an artificial visual method, and the problems of low efficiency, easy fatigue of workers, high omission rate, high false detection rate and the like exist. Some manufacturers also adopt an image recognition scheme to perform automatic detection, the scheme needs to collect the conical surface of the conical member first to obtain an image to be detected, and currently, an area-array camera is generally used for detecting the conical member, and the area-array camera shoots from a plurality of different angles to obtain an image sequence covering the whole conical surface. The method has the advantages of low detection efficiency, large data volume, high image splicing difficulty and difficult guarantee of splicing precision. If the linear array camera is used for acquiring the image of the conical member, the linear speed of the conical member in the rotating process of the conical member at different positions of the bus on the conical surface is different, so that the linear array camera has the characteristics of high large end and low small end, the problem that the large end (large radius and high linear speed) part is stretched and the small end (small radius and low linear speed) part is compressed exists in the acquired image, and the image is obviously distorted. For this reason, in the existing scheme, a complex mechanical device (such as a synchronous speed-changing mechanism) or software (such as dynamically adjusting the line frequency) is adopted to try to eliminate the image distortion in the process of capturing the image by the linear array camera, however, in practical application, such a system has complex design, high cost and low reliability. Disclosure of Invention The present application aims to solve one of the technical problems in the related art to a certain extent. To this end, the application provides a surface image acquisition method and a surface image acquisition device for a tapered member. In order to achieve the above purpose, the application adopts the following technical scheme that the surface image acquisition method for the conical member comprises the following steps: the linear array camera in the image acquisition unit is matched with the telecentric lens to perform line scanning on the rotary cone-shaped component at a fixed line frequency, so as to obtain an original image with regular distortion; Establishing a coordinate mapping relation between pixel coordinates of the original image and surface geometric coordinates of the conical member according to the fixed line frequency and the rotary motion of the conical member; And correcting the original image according to the coordinate mapping relation to obtain a target image. The application has the following beneficial effects that the linear scanning is carried out on the conical component which rotates at a fixed line frequency through the linear array camera in the image acquisition unit and the telecentric lens, so that the rotary scanning image with constant magnification and no perspective distortion can be acquired, namely, the acquired original image has the distortion problem, but the distortion is regular, and the subsequent algorithm correction is convenient. And establishing a coordinate mapping relation, correcting the acquired original image according to the coordinate mapping relation, and realizing pixel-level inverse mapping correction based on the cone geometric model to obtain a target image, wherein the target image can be used for an image recognition scheme to detect surface defects. Compared with the t