A method and system for detecting defects in a specularly reflecting object having a free-form surface
By using asymmetric pattern calibration and two-dimensional curvature formulas, the problem of detecting concave and convex defects in free-form surface mirror reflectors was solved, achieving high-precision defect detection results.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- XIAMEN WEIXINTAI TECH CO LTD
- Filing Date
- 2026-03-26
- Publication Date
- 2026-06-19
AI Technical Summary
Existing technologies are insufficient for effectively detecting uneven defects in specular reflective objects with free-form surfaces, as the curvature variations of free-form surfaces interfere with the detection results of traditional methods.
The curvature of the surface of a specular reflective object is calibrated by using an asymmetrical patterned light source. An image deviating from the grayscale is acquired by a line scan camera. By combining the two-dimensional curvature formula and the compensation pattern boundary, the matching and defect detection of freeform surfaces can be achieved.
It enables accurate defect detection of freeform mirror reflective objects, and can identify concave and convex defects in areas of curvature variation, thus improving the accuracy and reliability of detection.
Smart Images
Figure CN122238211A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of defect detection, and specifically to a defect detection method and defect inspection system for a specular reflective object with a free-form surface. Background Technology
[0002] In the inspection of defects on the surface of mirror-reflective objects, when the product itself is flat and has no curvature variation, defects can be directly quantified using black and white stripes of equal width. For example, when a white stripe with a grayscale value of 255 and a black stripe with a grayscale value of 0 enter the camera, the light intensity will be combined to a grayscale value of 128. When the camera captures a concave or convex defect, a darker grayscale value will be displayed on top of the background grayscale value of 128, thus identifying the corresponding defect. However, when the product is a free-form surface, it has its own curvature variation, such as metal parts like watch cases. In this case, the curvature of the free-form surface greatly interferes with the detection of concave or convex defects, and the above method cannot directly detect the defects. Summary of the Invention
[0003] The purpose of this invention is to overcome the aforementioned defects or problems existing in the prior art or to provide a material basis for overcoming the aforementioned defects or problems existing in the prior art, and to provide a defect detection method and defect inspection system for a specular reflective object with a free-form surface.
[0004] To achieve the above objectives, the present invention and its preferred embodiments employ the following technical solutions, but the embodiments are not limited to the following solutions: Option 1, a defect detection method for specular reflective objects with freeform surfaces, includes the following steps: Focusing on the light source, the light emitted by the light source is reflected by the specular reflector and received by a line scan camera with a telecentric lens. In this step, the curvature of the surface of the specular reflector is calibrated using an asymmetrical pattern. Move the mirror reflector so that its freeform surface is fully calibrated; Focusing on the specular reflector, the light source fits the stripe boundary of the freeform surface shape to match the freeform surface of the specular reflector. When there is an anomaly in the area of the freeform surface, the line scan camera acquires an image deviating from the first gray level at the anomaly location. When there is no anomaly in the area of the freeform surface, the line scan camera acquires an image of the first gray level. The mirror reflector is moved so that its entire free-form surface is detected.
[0005] Option 2, based on Option 1, the two-dimensional interval calculation formula for the freeform surface is: κ(t) = |x′(t) y″(t) y′(t) x″(t)| / [ (x′(t)^2 + y′(t)^2)^(3 / 2) ]; t: Curve parameter, which can represent the sampling number, scan position, or arc length parameter on the fringe boundary; x(t): The position function of the stripe boundary curve in the horizontal coordinate direction; y(t): The position function of the stripe boundary curve in the longitudinal coordinate direction; κ(t): The magnitude of the local curvature at the corresponding sampling point, used to characterize the degree of curvature of the fringe boundary; ; ; , This represents the standard pattern boundary position when there is no curvature and no defects. α x (t), α y (t) represents the offset after being affected by the freeform surface and local defects.
[0006] Option 3, based on Option 1, includes a number of LED beads arranged in an array and independently controllable to turn on and off.
[0007] Option 4, based on Option 3, is an LCD screen or an array of controllable light sources.
[0008] Option 5, based on Option 1, assumes the coordinates of a sampling point on the boundary of the projected pattern under the state of no curvature reference are P0(t)=(x0(t),y0(t)), and the coordinates of the boundary point collected under the action of the freeform surface are P(t)=(x(t),y(t)). Then the lateral offset α x (t)=x(t)-x0(t), longitudinal offset α y (t) = y(t) - y0(t); the corresponding compensation pattern boundary can be set as x c (t)=x0(t)-λ x α x (t), y c (t)=y0(t)-λ y α y (t), where λ x , λ y The compensation coefficient is used to restore the reflected image of the normal freeform surface area to the target reference gray level by loading the compensated boundary coordinates onto the LCD screen or an array of controllable light sources.
[0009] Option 6, a defect inspection system suitable for performing a defect detection method for a specular reflective object with a free-form surface as described in any one of Options 1 to 5, comprising: Line scan camera, which has a telecentric lens and is suitable for acquiring images; A light source, which is suitable for emitting light; A stage adapted to carry an object and to move relative to the line scan camera and the light source.
[0010] As can be seen from the above description of the present invention and its preferred embodiments, compared with the prior art, the technical solution of the present invention and its preferred embodiments have the following beneficial effects due to the adoption of the following technical means: A method for detecting defects in a specular reflective object with a free-form surface includes the following steps: Focusing on the light source, the light emitted by the light source is reflected by the specular reflector and received by a line scan camera with a telecentric lens. In this step, the light source uses an asymmetrical pattern to calibrate the curvature of the specular reflector surface. The specular reflector is moved so that the free surface of the specular reflector is fully calibrated. Thus, the state of the free surface is obtained by the distortion of the corresponding x and y axes on the image, and the free surface of the specular reflector is determined. The light source is focused on the specular reflector, and its fringe boundary is fitted to the shape of a freeform surface to match the freeform surface of the specular reflector. When anomalies exist in a region of the freeform surface, the line scan camera acquires an image deviating from the first gray level at the anomalous area; when there are no anomalies in a region of the freeform surface, the line scan camera acquires an image at the first gray level. Through fitting, the influence of the freeform surface is removed. When a defect is detected, it can be reflected by the abnormal gray value, thus detecting the defect. The specular reflector is moved so that the entire freeform surface of the specular reflector is detected. By moving the entire freeform surface, the result is obtained. This method, by first calibrating the surface curvature and then continuously changing the fringe shape to adapt to the freeform surface, completes the detection of unevenness defects on the surface of products with large curvature. Attached Figure Description
[0011] To more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings used in the following description of the embodiments are briefly introduced. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.
[0012] Figure 1 This is a schematic diagram of the line scanning and mapping platform in Example 1; Figure 2 This is a schematic diagram of the line scan pattern in Example 1; Figure 3 This is a schematic diagram of the defect inspection system for freeform surface products in Example 1; Figure 4 This is a schematic diagram of defect detection imaging in Example 1; Figure 5 This is a schematic diagram of the defect inspection system for products with flat surfaces in Example 1; Figure 6 This is the calculation formula for the freeform surface in Example 1. Detailed Implementation
[0013] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are preferred embodiments of the present invention and should not be considered as excluding other embodiments. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.
[0014] Unless otherwise expressly defined, the use of terms such as "first," "second," or "third" in the claims, description, and accompanying drawings of this invention is for distinguishing different objects and not for describing a specific order.
[0015] Unless otherwise expressly defined, in the claims, description, and accompanying drawings of this invention, the use of directional terms such as "center," "lateral," "longitudinal," "horizontal," "vertical," "top," "bottom," "inner," "outer," "upper," "lower," "front," "rear," "left," "right," "clockwise," and "counterclockwise" to indicate orientation or positional relationships is based on the orientation and positional relationships shown in the accompanying drawings and is only for the convenience of describing the invention and simplifying the description, and is not intended to indicate or imply that the device or element referred to must have a specific orientation or be constructed and operated in a specific orientation, and therefore should not be construed as limiting the specific scope of protection of this invention.
[0016] Unless otherwise expressly defined, the terms "fixed connection" or "fixed connection" used in the claims, description and drawings of this invention should be interpreted broadly to refer to any connection in which there is no displacement or relative rotation relationship between the two parties, including non-removable fixed connection, detachable fixed connection, integral connection and fixed connection by other means or components.
[0017] In the claims, description and accompanying drawings of this invention, the terms "comprising," "having," and variations thereof are used to mean "including but not limited to."
[0018] A method for detecting defects in a specular reflective object with a free-form surface includes the following steps: Focusing on the light source, the light emitted by the light source is reflected by the specular reflector and received by a line scan camera with a telecentric lens. In this step, the curvature of the specular reflector surface is calibrated using an asymmetrical pattern, with reference to... Figure 2Within the circumscribed rectangular area, a central horizontal dark band is set. Above and below this dark band, two bright areas defined by oblique boundaries are set respectively. The upper and lower bright areas are not mirror images of each other about the same vertical center line, but rather correspond about the central horizontal baseline, thus creating a corresponding asymmetrical pattern. When the irregular pattern scans the product, the width of each column of black or white stripes is standard and fixed. However, at the defect location, the stripes will deform. The horizontal direction represents the length, and the vertical direction represents the width. The lateral offset reflects the curvature change in the x-direction of the defect, denoted as αx; the vertical offset reflects the curvature change in the y-direction of the defect, denoted as αy. (Reference) Figure 6 The formula for calculating the two-dimensional interval of a freeform surface is: κ(t) = |x′(t) y″(t) y′(t) x″(t)| / [ (x′(t)^2 + y′(t)^2)^(3 / 2) ]; t: Curve parameter, which can represent the sampling number, scan position, or arc length parameter on the fringe boundary; x(t): The position function of the stripe boundary curve in the horizontal coordinate direction; y(t): The position function of the stripe boundary curve in the longitudinal coordinate direction; x′(t): The first derivative of x(t) with respect to parameter t, representing the rate of change of lateral position; y′(t): The first derivative of y(t) with respect to parameter t, representing the rate of change of longitudinal position; x′′(t): The second derivative of x(t) with respect to parameter t; y′′(t): The second derivative of y(t) with respect to parameter t; κ(t): The magnitude of the local curvature at the corresponding sampling point, used to characterize the degree of curvature of the fringe boundary; This formula characterizes the degree of geometric curvature of the fringe boundary curve in the image, and thus reflects the degree of reflection distortion caused by the free-form surface under test to the projected pattern.
[0019] Lateral offset αx represents the displacement of the stripe boundary point relative to the reference position in the x-direction; The longitudinal offset αy represents the displacement of the stripe boundary point relative to the reference position in the x-direction; x(t)=x0(t)+αx(t), y(t)=y0(t)+αy(t), x0(t) and y0(t) are the standard pattern boundary positions when there is no curvature and no defects; αx(t) and αy(t) are the offsets after being affected by the freeform surface and local defects; Substituting the offset x(t) and y(t) into the curvature formula above, we can obtain the curvature K(t) at the corresponding position. Therefore, αx and αy reflect the change in the position of the stripes; K(t) reflects the degree of local curvature further formed by this change in position.
[0020] The lateral and longitudinal offsets together determine the curve shape, and the curvature is a comprehensive characteristic calculated from this curve shape. In this application, the lateral and longitudinal offsets represent the displacement components of the fringe boundary points relative to the fringe boundary positions in the reference state in two orthogonal directions of the image coordinate system. Specifically, the lateral offset characterizes the amount of fringe boundary offset in the lateral direction, and the longitudinal offset characterizes the overall offset of the current reflection information at its corresponding longitudinal position in the original asymmetric pattern.
[0021] Specifically, in the line scanning scheme of this application, if the surface of the specular reflective object to be measured is an ideal plane, then with the relative positions of the light source, camera, and object to be measured fixed, the line scanning camera always receives the reflection information from a fixed position in the light source pattern during the scanning process, thus ultimately forming a stable striped image. At this time, the brightness ratio and the position of the black and white boundary in the local area correspond to the reference pattern.
[0022] When the surface under test has a free curvature change or local tilt, the direction of reflected light changes, and the position of the pattern actually received by the camera shifts relative to the reference state. Since this application uses an asymmetrical pattern, which has a non-uniform black-and-white width distribution or brightness-dark ratio distribution in the longitudinal direction, different longitudinal positions correspond to different local encoding rules. By comparing the local brightness-dark ratio rules acquired by the camera with the preset ratio rules in the reference pattern, the longitudinal position change corresponding to the current reflection information can be determined first, thereby obtaining the longitudinal offset. .
[0023] After determining the longitudinal position, the actual black and white boundary position acquired at this longitudinal position is compared with the corresponding boundary position under the reference state to obtain the displacement of the stripe boundary in the lateral direction, i.e., the lateral offset. .
[0024] Therefore, the lateral offset and the longitudinal offset are not simply equal, proportional, or algebraically related, but rather together constitute the two-dimensional offset of the fringe boundary point relative to the reference position. Let the coordinates of the fringe boundary point in the reference state be: P 0( t )=( x 0( t ), y 0( t )) The actual coordinates of the boundary points collected are:P ( t )=( x ( t ), y ( t )) Then we have: x(t) = (t)
[0025] y(t) = (t) +
[0026] Therefore, and These are the horizontal and vertical displacement components of the same boundary point relative to the reference position, which together determine the actual position of the boundary point in the image and its geometric shape changes.
[0027] The two-dimensional curvature formula given in this application specification is as follows: κ(t) = |x′(t) y″(t) y′(t) x″(t)| / [ (x′(t)^2 + y′(t)^2)^(3 / 2) ] This characterizes the local curvature of the fringe boundary curve. Here, x(t) and y(t) are position functions of the boundary point in the lateral and longitudinal directions, respectively, while the lateral and longitudinal offsets are precisely the displacements from the reference position to the actual position. In other words, the lateral and longitudinal offsets reflect the positional changes of the fringe boundary, and the curvature k(t) is a geometrically comprehensive characterization quantity further formed by this two-dimensional positional change.
[0028] Therefore, the longitudinal offset is used to determine the longitudinal corresponding position of the current reflection information in the asymmetric pattern, and the lateral offset is used to determine the lateral drift of the stripe boundary at that longitudinal position. Together, they determine the two-dimensional offset of the stripe boundary point, and can be further converted into the local slope information of the test point in two directions according to the system reflection geometry, so as to be used for free surface calibration and subsequent defect detection.
[0029] Move the mirror reflector so that its freeform surface is fully calibrated; Since the curvature calibration of freeform surfaces is based on establishing a curvature mapping relationship according to the overall reflection deformation of the surface under test caused by the projected asymmetric pattern, if the calibration area itself has concave or convex defects, these defects will introduce local additional offsets, making the acquired fringe distortion inaccurate. Therefore, it is necessary to ensure that the calibrated freeform surface is defect-free to obtain the correct free curvature and prevent it from affecting the detection.
[0030] Focusing on the specular reflector (the principle for selecting the focus position is to ensure that the reflected stripes within the target detection area meet the preset requirements for sharpness and grayscale stability, thereby guaranteeing the reliability of curvature calibration and defect detection), the light source is fitted with the fringe boundary of the freeform surface, matching it with the freeform surface of the specular reflector. This step can be achieved through the mapping relationship between the reference pattern coordinates, the offset, and the compensation pattern coordinates.
[0031] Let the coordinates of a sampling point on the boundary of the projected pattern under the condition of no curvature reference be P0(t)=(x0(t),y0(t)), and the coordinates of the boundary point collected under the action of the freeform surface be P(t)=(x(t),y(t)). Then the lateral offset α x (t)=x(t)-x0(t), longitudinal offset α y (t) = y(t) - y0(t). The corresponding compensation pattern boundary can be set as x. c (t)=x0(t)-λ x α x (t), y c (t)=y0(t)-λ y α y (t), where λ x , λ y The compensation coefficient is used. By loading the compensated boundary coordinates onto an LCD screen or an array of controllable light sources, the reflected image of a normal freeform surface area can be restored to the target reference gray level, while the local defect area still appears as an abnormal gray level, thus achieving defect detection.
[0032] (like Figure 3 (The corresponding curved surface area is no longer a 1:1 ratio of black and white stripes) When there is an anomaly in the area of the free surface, the line scan camera acquires an image deviating from the first gray level at the anomaly location. When there is no anomaly in the area of the free surface, the line scan camera acquires an image of the first gray level. In this embodiment, the first gray level is a gray value of 128.
[0033] By continuously changing the stripe shape and fitting the curvature of the freeform surface, uneven defects can also be presented on a base surface with a grayscale value of 128. Specifically, the stripe boundary of the freeform surface shape is fitted by a programmable trigger and a light source trigger to match the freeform surface. At this time, the deformed stripes received by the CCD camera relative to the projection image still represent the uneven defects on the surface of the object under test.
[0034] The mirror reflector is moved so that its entire free-form surface is detected.
[0035] The light source is an LCD screen, which includes several LED beads arranged in an array and independently controllable to turn on and off. This allows the light source to automatically change the lighting mode according to the freeform surface, thereby achieving matching with different freeform surfaces and enabling them to display images of the first grayscale under normal conditions.
[0036] A defect inspection system is suitable for performing defect detection methods on specular reflective objects with freeform surfaces as described above, comprising a line scan camera, a programmable trigger, and a light source controller.
[0037] Line scan cameras have telecentric lenses and are suitable for acquiring images; The light source is suitable for emitting light; The stage is suitable for carrying objects and for moving relative to the line scan camera and light source.
[0038] Programmable triggers are used to generate trigger signals according to preset rules, thereby achieving the corresponding work.
[0039] A light source controller is used to control the way a light source emits light.
[0040] The above-mentioned detection is achieved through the cooperation of various components, thereby completing the detection of concave and convex defects on the surface of a large curvature mirror reflector.
[0041] Compared with the prior art, this embodiment has the following beneficial effects: In one exemplary embodiment, a defect detection method for a specular reflective object with a freeform surface includes the following steps: Focusing on the light source, the light emitted by the light source is reflected by the specular reflector and received by a line scan camera with a telecentric lens. In this step, the light source uses an asymmetrical pattern to calibrate the curvature of the specular reflector surface. The specular reflector is moved so that the free surface of the specular reflector is fully calibrated. Thus, the state of the free surface is obtained by the distortion of the corresponding x and y axes on the image, and the free surface of the specular reflector is determined. The light source is focused on the specular reflector, and its fringe boundary is fitted to the shape of a freeform surface to match the freeform surface of the specular reflector. When anomalies exist in a region of the freeform surface, the line scan camera acquires an image deviating from the first gray level at the anomalous area; when there are no anomalies in a region of the freeform surface, the line scan camera acquires an image at the first gray level. Through fitting, the influence of the freeform surface is removed. When a defect is detected, it can be reflected by the abnormal gray value, thus detecting the defect. The specular reflector is moved so that the entire freeform surface of the specular reflector is detected. By moving the entire freeform surface, the result is obtained. This method, by first calibrating the surface curvature and then continuously changing the fringe shape to adapt to the freeform surface, completes the detection of unevenness defects on the surface of products with large curvature.
[0042] In one exemplary embodiment, the light source includes a plurality of LED beads arranged in an array and independently controllable to turn on and off, so as to automatically change the corresponding lighting mode according to the freeform surface, thereby achieving matching for different freeform surfaces and enabling them to present an image of the first grayscale under normal conditions.
[0043] The foregoing description of the specifications and embodiments is intended to explain the scope of protection of this invention, but does not constitute a limitation on the scope of protection of this invention. Modifications, equivalent substitutions, or other improvements to the embodiments of this invention or a portion thereof that can be obtained by those skilled in the art through logical analysis, reasoning, or limited experimentation, based on the teachings of this invention or the foregoing embodiments, in conjunction with common knowledge, general technical knowledge, and / or existing technology, should all be included within the scope of protection of this invention.
Claims
1. A method for detecting defects in a specular reflective object with a free-form surface, characterized in that: Includes the following steps: Focusing on the light source, the light emitted by the light source is reflected by the specular reflector and received by a line scan camera with a telecentric lens. In this step, the curvature of the surface of the specular reflector is calibrated using an asymmetrical pattern. Move the mirror reflector so that its freeform surface is fully calibrated; Focusing on the specular reflector, the light source fits the stripe boundary of the freeform surface shape to match the freeform surface of the specular reflector. When there is an anomaly in the area of the freeform surface, the line scan camera acquires an image deviating from the first gray level at the anomaly location. When there is no anomaly in the area of the freeform surface, the line scan camera acquires an image of the first gray level. The mirror reflector is moved so that its entire free-form surface is detected.
2. The defect detection method for a specular reflective object with a free-form surface as described in claim 1, characterized in that: The formula for calculating the two-dimensional interval of the freeform surface is: κ(t) = |x′(t) y″(t) y′(t) x″(t)| / [ (x′(t)^2 + y′(t)^2)^(3 / 2) ]; t: Curve parameter, which can represent the sampling number, scan position, or arc length parameter on the stripe boundary; x(t): The position function of the stripe boundary curve in the horizontal coordinate direction; y(t): The position function of the stripe boundary curve in the longitudinal coordinate direction; κ(t): The magnitude of the local curvature at the corresponding sampling point, used to characterize the degree of curvature of the fringe boundary; ； ； , This represents the standard pattern boundary position when there is no curvature and no defects. α x (t), α y (t) represents the offset after being affected by the freeform surface and local defects.
3. The defect detection method for a specular reflective object with a free-form surface as described in claim 1, characterized in that: The light source includes several LED beads arranged in an array and each of which can be independently controlled to turn on and off.
4. The defect detection method for a specular reflective object with a free-form surface as described in claim 3, characterized in that: The light source is an LCD screen or an array of controllable light sources.
5. The defect detection method for a specular reflective object with a free-form surface as described in claim 1, characterized in that: Let the coordinates of a sampling point on the boundary of the projected pattern under the condition of no curvature reference be P0(t)=(x0(t),y0(t)), and the coordinates of the boundary point collected under the action of the freeform surface be P(t)=(x(t),y(t)). Then the lateral offset α x (t)=x(t)-x0(t), longitudinal offset α y (t) = y(t) - y0(t); the corresponding compensation pattern boundary can be set as x c (t)=x0(t)-λ x α x (t), y c (t)=y0(t)-λ y α y (t), where λ x , λ y The compensation coefficient is used to restore the reflected image of the normal freeform surface area to the target reference gray level by loading the compensated boundary coordinates onto the LCD screen or an array of controllable light sources.
6. A defect inspection system, characterized in that: It is suitable for performing a defect detection method for a specular reflective object with a free-form surface as described in any one of claims 1-5, comprising: Line scan camera, which has a telecentric lens and is suitable for acquiring images; A light source, which is suitable for emitting light; A stage adapted to carry an object and to move relative to the line scan camera and the light source.