Method and system for calculating telecentricity based on light cone center line method

Abstract

The invention discloses a telecentricity calculation method and system based on a light cone center line method, and the method comprises the steps: placing a target plate containing a mark point in a light path, enabling the target plate to form a light spot on a target surface of an image sensor after the target plate is imaged through a to-be-measured optical assembly; collecting one-dimensional gray scale section images of the light spots in the X-axis direction and the Y-axis direction on a plurality of planes with different heights along the optical axis direction; performing edge detection on the one-dimensional gray scale section image on each height plane to obtain edge position information of a light spot in a corresponding direction; calculating the center coordinates of the light spot on each height plane in the X-axis direction and the Y-axis direction by adopting a gray scale weighted centroid method; based on each center coordinate, carrying out space straight line fitting to obtain a light cone center line; and calculating the telecentricity of the optical assembly to be measured in the X-axis and Y-axis directions according to the inclination angle of the center line of the light cone relative to the optical axis. According to the invention, efficient and accurate measurement of the telecentricity of the optical assembly can be realized, the cost is low, the precision is reliable, and the method is suitable for rapid on-line detection of a production field.

Description

Technical Field

[0001] This invention relates to the field of computer vision and image measurement technology, specifically a method and system for measuring the telecentricity of an optical system by finding the edge and calculating the centroid of a one-dimensional grayscale profile and fitting a three-dimensional light cone centerline. Background Technology

[0002] In an ideal telecentric optical system, both the object-side and image-side principal rays are parallel to the optical axis, and the system magnification is independent of object distance and image distance. Theoretically, a telecentric system should possess measurement stability that ignores changes in object distance. In reality, due to inherent deviations in optical design, manufacturing, and assembly processes, any telecentric lens has a non-negligible telecentricity error. This error causes a systematic drift in the size of the image on the sensor when the measured object moves vertically within the system's depth of field, introducing a non-linear measurement error that is difficult to compensate for simply with software. Telecentricity is not only a property of the lens itself but also a comprehensive performance characteristic of the entire imaging system. If the mounting plane of the camera sensor is not strictly perpendicular to the lens's optical axis, the effective telecentricity of the entire system will be significantly degraded.

[0003] Currently, accurate telecentricity measurement typically relies on specialized optical laboratories equipped with high-precision rotary stages. This equipment is expensive, complex to operate, and inefficient, making it unsuitable for quality control in production environments or system performance diagnostics at user sites. Therefore, the industry urgently needs a telecentricity assessment solution that is easy to operate, low-cost, accurate, and suitable for online or field measurements. Summary of the Invention

[0004] The technical problem to be solved by the present invention is to provide a method and system for calculating telecentricity based on the optical cone centerline method to address the shortcomings of the prior art. This method does not require a high-precision rotary stage, but only uses a conventional displacement stage and image sensor. By analyzing the changes in the center coordinates of the target image spot at multiple axial positions, the optical cone centerline is fitted and its spatial tilt angle is calculated, thereby achieving efficient and accurate measurement of the telecentricity of optical components or complete imaging systems. It is simple to operate, low in cost, and reliable in accuracy, and is suitable for rapid online detection in production sites.

[0005] To achieve the above-mentioned technical objectives, the technical solution adopted by the present invention is as follows:

[0006] A method for calculating telecentricity based on the optical cone centerline method includes the following steps:

[0007] (1) Target plate setting: Place the target plate containing the marker points in the optical path, that is, place it on the image side of the optical component under test. Adjust the initial position of the target plate so that after the marker points are imaged by the optical component under test, a clear and full circular light spot is formed on the target surface of the image sensor.

[0008] (2) Data acquisition: Along the optical axis of the optical component under test, one-dimensional grayscale profile images of the light spot in the mutually perpendicular X-axis and Y-axis directions are acquired on multiple planes at different heights;

[0009] (3) Edge detection: Edge detection is performed on the one-dimensional grayscale profile images in the X-axis and Y-axis directions collected on each height plane to obtain the edge position information of the light spot in the corresponding direction;

[0010] (4) Center positioning: Based on the edge position information, the gray-weighted centroid method is used to calculate the center coordinates of the light spot on each height plane in the X-axis and Y-axis directions respectively;

[0011] (5) Centerline fitting: Based on the center coordinates on each height plane, perform spatial straight line fitting to obtain the centerline of the light cone in three-dimensional space;

[0012] (6) Telecentricity calculation: Based on the tilt angle of the optical cone centerline relative to the optical axis, the telecentricity of the optical component under test in the X-axis and Y-axis directions is calculated.

[0013] As a further improved technical solution of the present invention, in step (1), the target plate is a calibration plate with a white background and a central black dot. The diameter of the light spot formed by the black dot through the optical component under test is larger than the smallest resolvable unit of the image sensor, and the signal-to-noise ratio of the light spot edge detection is greater than 20 dB.

[0014] As a further improvement of the present invention, in step (2), the target plate is moved at equal intervals along the optical axis (defined as the Z-axis) of the optical component under test by a precision displacement stage. At each stopped and stabilized height position, the image sensor is triggered to acquire one-dimensional grayscale profile images of the light spot in the X-axis and Y-axis directions, respectively. The X-axis and Y-axis correspond to the row and column directions that are perpendicular to each other on the sensor target surface.

[0015] As a further improvement of the present invention, in step (3), edge detection is performed on the one-dimensional profile image acquired at each height position (preferably using the Canny algorithm). The one-dimensional grayscale profile image is processed using the Canny edge detection algorithm, specifically including: Gaussian filtering for noise reduction, calculation of gradient magnitude and direction, non-maximum suppression, and determination of true edges using the double threshold method.

[0016] As a further improvement of the present invention, in step (4), within the detected edge contour of the light spot, the center coordinates of the light spot in the X-axis and Y-axis directions are calculated using the gray-scale weighted centroid method. The calculation formula is as follows:

[0017] (1);

[0018] (2);

[0019] in, The coordinates of the pixels within the edge contour. , Indicates the index of the pixel within the edge contour. For pixels The grayscale values ​​are summed over a range of all pixels within the detected edge contour; the calculated values ​​are... This refers to the centroid coordinates of the light spot on the corresponding height plane in the two-dimensional plane, which is also the center position coordinates of the light spot in the X-axis and Y-axis directions on the corresponding height plane, serving as the center point of the light cone in this cross section.

[0020] As a further improvement of the present invention, step (5) specifically includes: collecting the center coordinates (centroid coordinates) on all N height planes to form a three-dimensional point set {(X i ,Y i Z i Let the point set be ∣i=1,2,...,N}. First, robust estimation algorithms such as RANSAC are used to iteratively fit the point set to identify and remove outliers caused by dust, noise, or local defects. Then, the least squares method is applied to fit a three-dimensional spatial line to the valid point set after outlier removal; the resulting line is the spatial light cone centerline describing the direction of light cone propagation.

[0021] As a further improvement of the present invention, step (6) specifically includes:

[0022] The fitted centerline of the optical cone is projected onto a plane containing the optical axis and the X direction (i.e., the XZ plane) and a plane containing the optical axis and the Y direction (i.e., the YZ plane); the angle between the centerline and the Z-axis (ideal optical axis) in the corresponding projection plane is calculated. The telecentricity value is obtained by the following formula:

[0023] Through formula Calculate the telecentricity of the optical component under test in the X-axis direction;

[0024] Through formula Calculate the telecentricity of the optical component under test in the Y-axis direction;

[0025] in, The displacement of the center coordinates along the X-axis with height is the amount of drift of the spot center along the X-direction on the image plane (the pixel value needs to be converted into physical length according to the pixel size of the image sensor). The displacement along the Y-axis of the center coordinate system varies with height, i.e., the drift of the spot center along the Y-direction on the image plane (the pixel value needs to be converted into physical length according to the pixel size of the image sensor). This refers to the change in height, which is the actual physical displacement of the target plate along the optical axis.

[0026] To achieve the above-mentioned technical objectives, another technical solution adopted by the present invention is as follows:

[0027] An optical system telecentricity measurement system for implementing a telecentricity calculation method includes: an illuminator, a beam splitter, an optical component under test, a target plate containing marker points, a precision Z-axis displacement stage for carrying and moving the target plate along the optical axis, an image sensor (camera), and a processing unit.

[0028] The light emitted by the illuminator is directed perpendicularly to the beam splitter. The beam splitter reflects most of the incident light at 90 degrees (a small portion is transmitted downwards (waste light), which is absorbed by the light-absorbing paper), causing it to be redirected to the right horizontal direction. The light in the right horizontal direction then passes through the optical component under test and illuminates the marked point on the target board perpendicularly. After the target board is illuminated, its reflected light carries the spot information and returns along the original path, i.e., horizontally to the left, passing through the optical component under test again. Then, it reaches the beam splitter again, which transmits most of the horizontally to the left light (a small portion is reflected upwards (weak stray light)). Finally, the transmitted light is received by the image sensor to form a spot image for analysis.

[0029] The processing unit is used to perform data acquisition, edge detection, center localization, centerline fitting, and telecentricity calculation.

[0030] As a further improvement of the present invention, the optical component to be tested is a microscope objective or a telecentric lens, etc.

[0031] The beneficial effects of this invention are as follows:

[0032] 1. Advanced measurement principle and high precision: This invention defines telecentrism by tracing the true propagation axis (centerline) of the light cone in three-dimensional space, with clear physical meaning. Employing sub-pixel grayscale weighted centroid positioning and robust RANSAC fitting, it effectively suppresses the effects of noise, diffraction, and outliers, achieving highly repeatable and accurate measurements.

[0033] 2. Simple System, Low Cost: This invention completely eliminates the need for expensive and bulky high-precision turntables. The core of the measurement system consists of a standard camera (image sensor), a common precision displacement stage, and conventional optical components, resulting in low setup costs and ease of maintenance and widespread adoption. This invention requires only a standard camera and a common precision displacement stage to complete the measurement. By switching the position of the black dot on the target board, it can quickly measure multiple key positions such as the center and four corners, providing a full-field telecentric distribution map for the optical system and supporting rapid optimization, such as by adjusting the fast-turning mirror in the illuminator or projection lens group to optimize telecentricity.

[0034] 3. Simple operation and extremely high efficiency: This invention only needs to process one-dimensional grayscale profiles along the X / Y axes, rather than two-dimensional images. Two-dimensional image processing requires processing a one-dimensional array within the range of min(M,N) to max(M,N), where M represents the pixel length of the two-dimensional image, N represents the pixel width of the two-dimensional image, min(M,N) represents the minimum value of M and N, and max(M,N) represents the maximum value of M and N. This invention only needs to process a one-dimensional array of 2H (H represents the number of height planes, usually 11, much smaller than min(M,N)), thus reducing the amount of computation, resulting in a very small data volume and a fast algorithm speed. Single-point measurement can be completed within tens of seconds, making it very suitable for full inspection or rapid sampling on production lines. Attached Figure Description

[0035] Figure 1 This is a schematic diagram of the detection system of the present invention.

[0036] Figure 2 This is a schematic diagram of the light cone.

[0037] Figure 3 This is a schematic diagram of the detection principle. Detailed Implementation

[0038] The specific embodiments of the present invention will be further described below with reference to the accompanying drawings:

[0039] This invention provides a telecentricity testing system based on the optical cone centerline method, which is constructed as follows:

[0040] like Figure 1 As shown, the system mainly includes an illuminator A1, a beam splitter L1, an optical component under test (such as a microscope C1), a white target plate D1 with a central black dot, an image sensor (integrated into a camera B1), a precision Z-axis displacement stage (not shown in the figure), and a processing unit. The beam splitter L1 reflects the incident illumination light to the right onto the target plate D1 (a small portion of the downward transmitted light on the beam splitter L1 is absorbed by a light-absorbing material to eliminate stray light interference), and transmits the reflected light from the target plate D1 to the camera B1, forming a circular light spot on the target surface of the image sensor in the camera B1.

[0041] Based on the above system, this invention provides a method for calculating telecentricity based on the optical cone centerline method, comprising the following steps:

[0042] (1) For example Figure 1 As shown, a target plate D1 is prepared. The target plate D1 is a white background target plate containing a black dot at the center. It is placed on the image plane of the lens. The diameter of the black dot of the target plate D1 after imaging by the optical system is larger than the smallest resolvable unit of the camera to ensure that the edge detection signal-to-noise ratio is >20dB. The optical system refers to the combination of optical devices that can capture or transmit light and form a visual image, including a beam splitter L1, the optical component under test (such as a microscope C1), the target plate D1, and an image sensor.

[0043] (2) Illuminator A1 in the optical system is activated to illuminate beam splitter L1. The beam splitter reflects most of the incident light at 90 degrees and transmits a small portion of the incident light downwards (the transmitted light will be absorbed by the light-absorbing paper), causing it to turn to the right horizontal direction. The light in the right horizontal direction then passes through the optical component under test (i.e., microscope C1) and vertically illuminates the marked point on the target plate D1. After the target plate D1 is illuminated, its reflected light carries the light spot information and returns along the original path, i.e., horizontally to the left, and passes through the optical component under test (i.e., microscope C1) again. Then, it reaches beam splitter L1 again. Beam splitter L1 transmits most of the horizontally to the left light. Finally, the transmitted light is received by camera B1 and forms a circular light spot on the target surface of the image sensor of camera B1.

[0044] (3) Control the precision Z-axis displacement stage to move the target plate D1 at equal intervals. After each movement, stabilize for 3 seconds (to eliminate mechanical vibration) and trigger the camera B1 to acquire the one-dimensional grayscale profile of the X / Y axis (the exposure time is adaptively adjusted to ensure that the saturation of the light spot meets 80%~90%). Control the camera B1 to acquire the one-dimensional grayscale profile images of the light spot in the X-axis and Y-axis directions on multiple preset planes at different heights.

[0045] (4) Edge detection is performed on the grayscale profile image of each height plane using the Canny algorithm: First, the image is denoised by Gaussian filtering, then the gradient magnitude and direction are calculated, the edges are thinned by non-maximum suppression, and finally the real edges are determined by the double threshold method to obtain the boundary contours of the light spot in the X and Y axis directions.

[0046] (5) For example Figure 2 As shown, the centroid of the spot boundary is calculated based on the edge detection results, and this centroid is taken as the midpoint of the X-axis and Y-axis boundaries at this height level. The centroid coordinates in the X-axis direction and the Y-axis direction are calculated by statistically analyzing the gray-level weighted coordinates of the pixels within the edge contour. , ,in The pixel coordinates are within the edge contour. , for The grayscale value of a pixel. The index of the pixel within the edge contour is used; the summation range is all pixels within the detected edge contour; the calculated value is... The coordinates of the centroid of the light spot on the height plane in the X and Y directions (that is, the X / Y coordinates of the center of the light spot on the height plane);

[0047] (6) Repeat steps (4)-(5) to obtain the boundary centroids on all height planes; Figure 2 P1, P2, and P3 are the centroids (i.e., center coordinates) of the light spot boundary calculated using the above method at different height planes (Z-axis positions).

[0048] (7) Outlier removal using RANSAC algorithm: Iteratively fit the boundary centroid coordinates of all height planes to remove outlier centroids. Use the least squares method to fit all retained boundary centroids to obtain the light cone centerline, for example... Figure 2 Connect P1, P2, and P3 in the middle;

[0049] (8) For example Figure 3 As shown, calculate the center line of the light cone and the optical axis (e.g. Figure 3 The angle between the Z-axis and the Z-axis and This refers to the telecentricity of the optical component under test: the center line of the light cone (such as the line connecting P1, P2, and P3) is projected onto the YZ plane respectively (the resulting lines are...). Figure 3 The line passing through P5) and the XZ plane (the line obtained after projection is) Figure 3 On the line passing through P4, through the formula and Calculate the centroidality in the X and Y axes respectively. and , , The displacement of the midpoint (i.e., the midpoint of the x-axis and y-axis boundaries in step (5)) in the X and Y directions. This refers to the change in altitude;

[0050] This embodiment obtains the telecentricity by measuring and analyzing the center coordinates of a series of light spots to fit and calculate the centerline of the light cone and its spatial tilt angle.

[0051] The scope of protection of this invention includes, but is not limited to, the above embodiments. The scope of protection of this invention is defined by the claims. Any substitutions, modifications, or improvements to this technology that are easily conceived by those skilled in the art fall within the scope of protection of this invention.

Claims

1. A method for calculating telecentricity based on the optical cone centerline method, characterized in that, Includes the following steps: (1) Target plate setting: The target plate containing the marker points is set in the optical path so that after being imaged by the optical component under test, it forms a light spot on the target surface of the image sensor; (2) Data acquisition: Along the optical axis of the optical component under test, one-dimensional grayscale profile images of the light spot in the mutually perpendicular X-axis and Y-axis directions are acquired on multiple planes at different heights; (3) Edge detection: Edge detection is performed on the one-dimensional grayscale profile images in the X and Y directions collected on each height plane to obtain the edge position information of the light spot in the X and Y directions; (4) Center positioning: Based on the edge position information, the gray-weighted centroid method is used to calculate the center coordinates of the light spot on each height plane in the X-axis and Y-axis directions respectively; (5) Centerline fitting: Based on the center coordinates on each height plane, spatial straight line fitting is performed to obtain the centerline of the light cone in three-dimensional space; (6) Telecentricity calculation: Based on the tilt angle of the optical cone centerline relative to the optical axis, the telecentricity of the optical component under test in the X-axis and Y-axis directions is calculated.

2. The method for calculating telecentricity based on the optical cone centerline method according to claim 1, characterized in that, In step (1), the target plate is a calibration plate with a white background and a central black dot. The diameter of the light spot formed by the black dot through the optical component under test is larger than the smallest resolvable unit of the image sensor, and the signal-to-noise ratio of the light spot edge detection is greater than 20dB.

3. The method for calculating telecentricity based on the optical cone centerline method according to claim 1, characterized in that, In step (2), the target plate is moved at equal intervals along the optical axis by a precision displacement stage to achieve image acquisition on multiple planes at different heights.

4. The method for calculating telecentricity based on the optical cone centerline method according to claim 1, characterized in that, In step (3), the Canny edge detection algorithm is used to process the one-dimensional grayscale profile image, specifically including: Gaussian filtering for noise reduction, calculation of gradient magnitude and direction, non-maximum suppression, and double thresholding to determine the real edge.

5. The method for calculating telecentricity based on the optical cone centerline method according to claim 1, characterized in that, In step (4), within the detected edge contour of the light spot, the center coordinates of the light spot in the X and Y axis directions are calculated using the gray-weighted centroid method. The calculation formula is as follows: （1）； （2）； in, The coordinates of the pixels within the edge contour. , Indicates the index of the pixel within the edge contour. For pixels The grayscale values ​​are summed over a range of all pixels within the detected edge contour; the calculated values ​​are... The coordinates of the center of the light spot on the height plane in the X and Y directions are given.

6. The method for calculating telecentricity based on the optical cone centerline method according to claim 1, characterized in that, Step (5) specifically includes: using the RANSAC algorithm to iteratively fit the center coordinates on all height planes to remove outliers; using the least squares method to perform spatial line fitting on the center coordinates retained after removing outliers to obtain the center line of the light cone in three-dimensional space.

7. The method for calculating telecentricity based on the optical cone centerline method according to claim 1, characterized in that, The specific steps (6) are as follows: The centerline of the light cone in three-dimensional space is projected onto a plane containing the light axis and the X-axis, and a plane containing the light axis and the Y-axis, respectively. Through formula Calculate the telecentricity of the optical component under test in the X-axis direction. ; Through formula Calculate the telecentricity of the optical component under test in the Y-axis direction. ; in, Let X be the displacement of the center coordinate along the X-axis as a function of height. Let be the displacement of the center coordinate along the Y-axis as a function of height. This represents the change in altitude.

8. A telecentricity measurement system for an optical system, used to implement the telecentricity calculation method as described in any one of claims 1-7, characterized in that, include: Illuminator, beam splitter, optical components under test, target plate containing markers, precision displacement stage for carrying and moving the target plate along the optical axis, image sensor and processing unit; The light emitted by the illuminator is directed perpendicularly to the beam splitter. The beam splitter reflects most of the incident light 90 degrees, turning it to the right horizontal direction. The light in the right horizontal direction then passes through the optical component under test and illuminates the marked point on the target board perpendicularly. After the target board is illuminated, its reflected light carries the spot information and returns along the original path, that is, horizontally to the left, passing through the optical component under test again. Then, it reaches the beam splitter again. The beam splitter transmits most of the horizontally to the left light. Finally, the transmitted light is received by the image sensor to form a spot image for analysis. The processing unit is used to perform data acquisition, edge detection, center localization, centerline fitting, and telecentricity calculation.

9. The optical system telecentricity measurement system according to claim 8, characterized in that, The optical component to be tested is a microscope.

Images