Method for correcting refractive index distortion in OCT (Optical Coherence Tomography)
By extracting surface point cloud data and normal vectors from OCT images, and using the three-dimensional refraction law and optical path update formula to correct the direction of light propagation, the refractive index distortion problem in OCT images was solved, and the true shape reconstruction of the sample was achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2025-10-29
- Publication Date
- 2026-03-13
AI Technical Summary
Image distortion caused by refractive index distortion in OCT images cannot accurately reflect the actual condition of the sample, thus affecting the accuracy of subsequent measurements.
By acquiring the original OCT image of the sample, surface point cloud data and normal vector information are extracted. The direction of light propagation is iteratively updated using the three-dimensional refraction law. The position of the corrected pixel is iteratively calculated according to the optical path update formula, the surface normal vector is reconstructed, and finally the distortion-corrected OCT image is output.
It achieves true shape reconstruction of OCT images, provides true scanning range information of the sample, and provides an accurate basis for subsequent measurements.
Smart Images
Figure CN121655374A_ABST
Abstract
Description
Technical Field
[0001] This invention relates to the field of optical coherence tomography (OCT) technology, and particularly to a method for correcting refractive index distortion in OCT. Background Technology
[0002] Optical coherence tomography (OCT) is a non-invasive optical imaging technique. It constructs high-resolution three-dimensional images by measuring the interference signals between the faint light signals reflected from different depths of the sample and the light reflected from a reference mirror. Its key lies in utilizing a low-coherence light source; only tissue-reflected light with a path matching the reference light source can interfere, thus achieving depth resolution.
[0003] When measuring a sample using an OCT device, a beam of light is incident perpendicularly from a point on the sample. The detector receives all backscattered / reflected signals along the light's propagation path, and after processing, obtains a scattering / reflectivity intensity distribution curve at that location. This one-dimensional data is called an aline, where the position of each point represents the optical path difference between the light and a reference surface at that point. If the sample being measured is a lens, the light's propagation direction changes after entering the first surface of the lens; therefore, the signal received by the detector is not strictly in the depth direction. However, when recovering the spatial domain information corresponding to the aline signal, the aline is generally treated as one-dimensional data, thus ignoring the changes in the light's propagation path in physical space. This results in distortion in the recovered spatial domain signal. Since this distortion is caused by the refractive index, it is called refractive index distortion.
[0004] By rapidly moving a scanning beam, multiple adjacent ails are continuously acquired along a straight line on the sample surface. Arranging these ails in spatial order creates a two-dimensional cross-sectional image, known as a BSCAN image. Similarly, combining different BSCAN two-dimensional images together forms a three-dimensional OCT image, revealing the sample's three-dimensional spatial distribution information.
[0005] Since the data collected in each aline does not come entirely from the depth direction of that point, and the distribution of the collected data is determined by the optical path of the light reaching that depth rather than the actual physical distance, the resulting three-dimensional OCT image cannot truly reflect the actual situation of the sample, and the refractive index distortion needs to be corrected. Summary of the Invention
[0006] The technical problem to be solved by the present invention is: in order to overcome the above-mentioned technical problem, the present invention provides a method for correcting refractive index distortion in OCT, which can remove the distortion caused by the refractive index of the sample and obtain image information within the true scanning range, so as to further measure the sample.
[0007] The technical solution adopted by this invention to solve its technical problem is: a method for correcting refractive index distortion in OCT, comprising the following steps: Step 1: Obtain the original OCT image of the sample to be tested, and extract the surface point cloud data and normal vector information of each layer; Step 2: Iteratively update the direction of light propagation according to the three-dimensional refraction law formula; Step 3: Iteratively calculate the corrected pixel position according to the optical path update formula; Step 4: Reconstruct the surface normal vector of the next layer based on the correction results of the current layer; Step 5: Output the OCT image after refractive index distortion correction.
[0008] In step 2, the formula for the three-dimensional law of refraction is: (1); In formula (1), the subscript i represents the position of the incident point of the ray. Let i be the direction of propagation of the incident light at incident position i. Let be the normal vector of the sample surface. The direction of propagation of the emitted light. Let be the refractive index of the medium in which the incident light is located. is the refractive index of the medium in which the emitted light is located.
[0009] In step 3, the optical path update formula is: (2); in Given the position of the point, The position of the corrected point. Let L be the direction of light propagation, and L be the distance the light travels between two points. If the refractive index of the medium is n, then L = OPL / n, where OPL (Optical Path Length) is the optical path length.
[0010] In step 4, the surface normal vector of the next layer is generated by the point The plane is obtained by fitting points nearby, and the normal vector of the fitted plane is the point. .
[0011] The optical path length (OPL) is obtained directly from the reconstructed OCT data. The OPL is calculated using the following formula: OPL = R × abs (L2 - L1), where R is the axial pixel resolution of the OCT system in μm / pixel; L2 is the position of the light ray exiting the sample in the aline to be corrected, L1 is the position of the light ray incident on the sample in the aline to be corrected in pixels; and abs is the absolute value.
[0012] Initial incident direction of light It is determined by the entire imaging system.
[0013] Assuming the incident ray is perpendicular to the sample at any position i, the initial incident direction of the ray... Most OCT systems are telecentric scanning systems, in which case the initial incident direction of the light can be considered to be perpendicular to the sample and all light rays are parallel to each other, which is the assumption here; for special OCT systems, the incident light rays can be determined according to the actual situation of the system.
[0014] The normal vector of the first surface of the sample is obtained directly from the point cloud data corresponding to the OCT image of the sample, while the normal vectors of the second and subsequent surfaces are determined by the new sample surface formed by the corrected light rays.
[0015] Normal vector of sample surface It is determined by the plane normal vector defined by the neighborhood of each point on the sample surface.
[0016] The beneficial effects of this invention are multi-layered. The method for correcting refractive index distortion in OCT mainly corrects the direction of light propagation through the optical refraction law and corrects the optical path through the sample refractive index. This method has a simple process, can obtain the true shape of the sample being tested, and lays the foundation for subsequent measurement of parameters such as distance and radius of curvature of the sample. It has a wide range of applications. Attached Figure Description
[0017] Figure 1 This is a flowchart of a method for correcting refractive index distortion in OCT according to the present invention.
[0018] Figure 2 This is a schematic diagram of the vector relationship in the three-dimensional law of refraction.
[0019] Figure 3 This is a schematic diagram of the correction result of a plano-concave lens using the correction method of the present invention, aline.
[0020] Figure 4 These are BSCAN images of a plano-convex lens before and after correction using the correction method of this invention. Figure (a) is the BSCAN image of the plano-convex lens before correction, and Figure (b) is the BSCAN image of the plano-convex lens after correction.
[0021] Figure 5 These are three-dimensional OCT images of a plano-convex lens before and after correction using the correction method of this invention. Figure (a) is the three-dimensional OCT image of the plano-convex lens before correction; Figure (b) is the three-dimensional OCT image of the plano-convex lens after correction. Detailed Implementation
[0022] The invention will now be described in further detail with reference to the accompanying drawings. It should be emphasized that the following description is merely exemplary and not intended to limit the scope or application of the invention.
[0023] A method for correcting refractive index distortion in OCT includes the following steps: Step 1: Obtain the original OCT image of the sample to be tested, and extract the surface point cloud data and normal vector information of each layer.
[0024] Step 2: Iteratively update the direction of light propagation according to the three-dimensional refraction law formula. Figure 2 This is a schematic diagram of the optical refraction law. Correspondingly, formula (1) is the optical refraction law in three-dimensional space: (1); In formula (1), the subscript i represents the position of the incident point of the ray. Let i be the direction of propagation of the incident light at incident position i. Let be the normal vector of the sample surface. The direction of propagation of the emitted light. Let be the refractive index of the medium in which the incident light is located. is the refractive index of the medium in which the emitted light is located.
[0025] Step 3: Iteratively calculate the corrected pixel position according to the optical path update formula: (2); in Given the position of the point, The position of the corrected point. Let L be the direction of light propagation, and L be the distance the light travels between the two points. If the refractive index of the medium is n, then L = OPL / n, where OPL is the optical path length.
[0026] In step 4, the surface normal vector of the next layer is generated by the point The plane is obtained by fitting points nearby, and the normal vector of the fitted plane is the point. Here, point Nearby points are those points. The radius is 3-5 pixels, centered at a point. This radius must be greater than the noise point spacing in the OCT image, but smaller than the local radius of curvature of the sample surface to ensure the stability of the fit. The fitting process involves using the least squares method to perform a plane fit on the nearby points.
[0027] The optical path length (OPL) is obtained directly from the reconstructed OCT data. The OPL is calculated using the following formula: OPL = R × abs (L2 - L1), where R is the axial pixel resolution of the OCT system in μm / pixel; L2 is the position of the light ray exiting the sample in the aline to be corrected, L1 is the position of the light ray incident on the sample in the aline to be corrected in pixels; and abs is the absolute value.
[0028] Initial incident direction of light It is determined by the entire imaging system.
[0029] Assume that the incident ray is perpendicular to the sample at any position i, i.e., the initial incident direction of the ray. .
[0030] The normal vector of the first surface of the sample is obtained directly from the point cloud data corresponding to the OCT image of the sample, while the normal vectors of the second and subsequent surfaces are determined by the new sample surface formed by the corrected light rays.
[0031] Normal vector of sample surface The plane normal vector determined by the neighborhood of each point on the sample surface is determined by a plane fitting method.
[0032] Step 4: Reconstruct the surface normal vector of the next layer based on the correction results of the current layer.
[0033] Step 5: Output the OCT image after refractive index distortion correction.
[0034] In OCT technology, the most basic data unit is called an aline (or A-scan), which is one-dimensional data acquired at a single spatial point, containing information about the intensity of reflected light along the depth direction of the sample. Multiple alines can be combined in space to form a two-dimensional cross-sectional image, and further combining multiple two-dimensional cross-sectional images yields a three-dimensional OCT image. However, because light is deflected in the sample due to changes in refractive index, traditional methods directly use the depth coordinates of the aline as the geometric depth for image reconstruction, leading to significant geometric distortion. The correction method of this invention is based on reverse ray tracing of each data point in each aline. The core steps of the correction method of this invention include data acquisition, direction update, position update, normal vector reconstruction, and result output. Figure 1 The specific algorithm implementation flow of a preferred embodiment of the present invention is shown below: First, the initial incident direction distribution of the light is obtained by measurement or calibration. Subsequently, the aline data in the OCT image is read one by one, and for each aline, each pixel is iteratively corrected using formulas (1) and (2). When all the points in an aline are corrected, the correction process of the next aline is started until all aline data is processed, and finally the corrected three-dimensional OCT image is output.
[0035] Example 1 In this embodiment, the sample is a plano-concave lens manufactured by Beijing Optoelectronics Technology Co., Ltd. The material is H-K9L, the measured refractive index is 1.52, and the lens thickness is 1.5±0.2mm. The measured thickness is 1526.693um, measured using spectral confocal measurement. After correction, the lens thickness is 1545.2um, with an error of 18.5um. The correction results for dozens of adjacent aline lenses are as follows: Figure 3 As shown in the figure. The red line represents the position of the first and second surfaces of the sample before correction, and the blue line represents the position of the first and second surfaces of the sample after correction. Figure 3 The uppermost position in the image represents the location of the first surface, and the lowermost position represents the location of the second surface. It can be observed that the first surfaces of the sample completely overlap before and after correction because the light propagates entirely in the air without deflection or change in refractive index. After passing through the first surface, the light is deflected, therefore the second surface does not overlap before and after correction. Furthermore, the distance between the two surfaces is shorter after correction than before correction because the refractive index of the sample is greater than that of air, resulting in a shorter actual physical spatial distance with the optical path remaining constant. All these phenomena are consistent with expectations.
[0036] Example 2 In this embodiment, the sample is a plano-convex lens manufactured by Beijing Optoelectronic Technology Co., Ltd. The material is H-K9L, the measured refractive index is 1.52, the lens thickness is 2.1±0.2mm, the measured thickness is 1998.418um, the measurement method is spectral confocal measurement, and the corrected lens thickness is 2009.3um with an error of 10.9um.
[0037] Figure 4 and Figure 5 In the test using a plano-convex lens, the image formed by the sample placement platform has been removed. It can be seen that the second surface of the corrected sample presents a true planar state, rather than a curved surface under distortion. Moreover, the distance between the two surfaces is consistent with the actual sample thickness, thus achieving the effect of distortion correction.
[0038] Based on the above-described preferred embodiments of the present invention, and through the foregoing description, those skilled in the art can make various changes and modifications without departing from the inventive concept. The technical scope of this invention is not limited to the contents of the specification, but must be determined according to the scope of the claims.
Claims
1. A method for correcting refractive index distortion in OCT, characterized in that, Includes the following steps: Step 1: Obtain the original OCT image of the sample to be tested, and extract the surface point cloud data and normal vector information of each layer; Step 2: Iteratively update the direction of light propagation according to the three-dimensional refraction law formula; Step 3: Iteratively calculate the corrected pixel position according to the optical path update formula; Step 4: Reconstruct the surface normal vector of the next layer based on the correction results of the current layer; Step 5: Output the OCT image after refractive index distortion correction.
2. The method for correcting refractive index distortion in OCT as described in claim 1, characterized in that, In step 2, the formula for the three-dimensional law of refraction is: (1); In formula (1), the subscript i represents the position of the incident point of the ray. Let i be the direction of propagation of the incident light at incident position i. Let be the normal vector of the sample surface. The direction of propagation of the emitted light. Let be the refractive index of the medium in which the incident light is located. is the refractive index of the medium in which the emitted light is located.
3. The method for correcting refractive index distortion in OCT as described in claim 2, characterized in that, In step 3, the optical path update formula is: (2); in Given the position of the point, The position of the corrected point. Let L be the direction of light propagation, and L be the distance the light travels between the two points. If the refractive index of the medium is n, then L = OPL / n, where OPL is the optical path length.
4. The method for correcting refractive index distortion in OCT as described in claim 3, characterized in that, In step 4, the surface normal vector of the next layer is generated by the point The plane is obtained by fitting points nearby, and the normal vector of the fitted plane is the point. .
5. The method for correcting refractive index distortion in OCT as described in claim 3, characterized in that, in, OPL is obtained directly from the reconstructed OCT data.
6. The method for correcting refractive index distortion in OCT as described in claim 5, characterized in that, The optical path length (OPL) is calculated using the following formula: OPL = R × abs (L2 - L1), where R is the axial pixel resolution of the OCT system, in μm / pixel; L2 is the position of the light ray exiting the sample from the aline to be corrected, L1 is the position of the light ray incident on the sample from the aline to be corrected, in pixels; and abs is the absolute value.
7. The method for correcting refractive index distortion in OCT as described in claim 3, characterized in that, Initial incident direction of light It is determined by the entire imaging system.
8. The method for correcting refractive index distortion in OCT as described in claim 7, characterized in that, Assuming the incident ray is perpendicular to the sample at any position i, the initial incident direction of the ray... .
9. The method for correcting refractive index distortion in OCT as described in claim 3, characterized in that, The normal vector of the first surface of the sample is obtained directly from the point cloud data corresponding to the OCT image of the sample, while the normal vectors of the second and subsequent surfaces are determined by the new sample surface formed by the corrected light rays.
10. The method for correcting refractive index distortion in OCT as described in claim 4, characterized in that, Normal vector of sample surface It is determined by the plane normal vector defined by the neighborhood of each point on the sample surface.