Refraction measurement system and method

Abstract

The invention provides a refraction measurement system and method, and the method comprises the steps: determining angular spectrum information related to the refraction attribute of a target sample through a sample image of each illumination angle in the illumination process of the target sample through a light-emitting diode array, carrying out the enhancement processing of the angular spectrum information in each sample image, and obtaining a feature enhancement image set; determining refraction interference characteristics of the target sample in the refraction measurement process, and performing gradient noise reduction on refraction interference of the target sample according to the refraction interference characteristics and the characteristic enhancement image set to obtain refraction phase gradient characteristics of the target sample after refraction noise reduction; determining quantitative phase distribution of refraction of the target sample according to a preset refraction characteristic quantity and a phase gradient characteristic of refraction of the target sample; and determining diopter distribution of the target sample according to the quantitative phase distribution and the refractive index and thickness of the target sample. By adopting the scheme of the invention, the influence of optical interference and noise on refraction measurement can be overcome under multi-angle illumination.

Description

Technical Field

[0001] This application relates to the field of measurement technology, and more specifically, to a refractive measurement system and method. Background Technology

[0002] Measurement is the process of quantifying and recording the physical properties or abstract characteristics of things through specific tools, standard units, and scientific methods. It transforms vague quantities, sizes, and strengths into specific values ​​in a repeatable and verifiable manner. It is not only a basic means of scientific research, but also widely used in engineering construction, manufacturing, and daily life, providing objective basis for decision-making, comparison, and improvement.

[0003] Refractive measurement is a core technology in ophthalmic diagnosis, optical correction, and the development of related devices. Its measurement accuracy directly affects clinical treatment outcomes and the performance of optical products. Currently, most mainstream refractive measurement technologies employ single-angle or limited-angle illumination. This illumination mode makes it difficult to comprehensively capture the refractive characteristic correlation information of different regions of the target sample, resulting in weak feature signals related to refractive attributes in the acquired sample images. At the same time, the measurement process is easily affected by various factors such as ambient light interference, sample surface reflection, and differences in the sample's own subtle structure, causing a large amount of noise to be mixed in the extracted refractive-related features. Traditional signal processing methods lack targeted processing mechanisms for these interfering features, making it difficult to achieve accurate noise reduction. Consequently, subsequent quantitative analysis of the sample's refractive state is biased and cannot accurately reflect the true refractive distribution of the target sample. Especially in scenarios with high requirements for high-precision measurement, the shortcomings of existing technologies are more prominent, making it difficult to meet the requirements of measurement accuracy and reliability in practical applications. Therefore, how to overcome the influence of optical interference and noise on refractive measurement under multi-angle illumination has become a problem facing the industry. Summary of the Invention

[0004] This application provides a refractive measurement system and method that can overcome the influence of optical interference and noise on refractive measurement under multi-angle illumination.

[0005] In a first aspect, this application provides a refractive measurement method, wherein a light-emitting diode array is used to illuminate the target sample of the refractive error to be measured from multiple angles. The method includes the following steps: Acquire sample images from various illumination angles during the process of illuminating the target sample with a light-emitting diode array; Angular spectral information related to the refractive properties of the target sample is extracted from each sample image, and the angular spectral information in each sample image is enhanced to obtain a feature-enhanced image set. Determine the refractive interference characteristics of the target sample during the refractive measurement process, and perform gradient denoising on the refractive interference of the target sample based on the refractive interference characteristics and the feature-enhanced image set to obtain the phase gradient characteristics of the refractive error of the target sample after refractive denoising. The quantitative phase distribution of the refractive index of the target sample is determined based on the preset refractive characteristic quantity of the target sample and the phase gradient characteristic of the refractive index. The refractive power distribution of the target sample is determined based on the quantitative phase distribution and the refractive index and thickness of the target sample.

[0006] In some embodiments, extracting spectral information related to the refractive properties of the target sample from each sample image specifically includes: Denoising is performed on each sample image to obtain each denoised sample image; Obtain the refractive properties of the target sample; Angular spectral information related to the refractive properties of the target sample is extracted from each denoised sample image.

[0007] In some embodiments, enhancing the angular spectral information in each sample image to obtain a feature-enhanced image set specifically includes: The angular spectral information in each sample image is enhanced to obtain the enhanced angular spectral information of each sample image; The feature-enhanced image set is determined based on the enhanced angular spectrum information in each sample image.

[0008] In some embodiments, determining the refractive interference characteristics of the target sample during the refractive measurement process specifically includes: Identify all interference sources in the target sample during refractive measurement. Determine the interference characteristics of each interference source; The refractive interference characteristics of the target sample during the refractive measurement process are determined based on all interference characteristics.

[0009] In some embodiments, performing gradient denoising on the refractive interference of the target sample based on the refractive interference features and the feature-enhanced image set to obtain the phase gradient features of the refractive error of the target sample after refractive denoising specifically includes: Select one feature-enhanced image from the feature-enhanced image set as the selected feature-enhanced image, and determine the initial phase gradient map of the selected feature-enhanced image; Based on the refractive interference characteristics, determine the effective gradient region and interference gradient region of the refractive field in the initial phase gradient map; The interference gradient region is subjected to interference suppression to obtain the interference gradient region after interference suppression; The effective gradient region is denoised to obtain the denoised effective gradient region. The phase gradient of the refractive image after refractive denoising is determined based on the interference gradient region after interference suppression and the effective gradient region after denoising. Continue to determine the phase gradient of the refractive image after refractive noise reduction in the image enhanced by remaining features; All phase gradients are used as the phase gradient features of the refractive image after refractive noise reduction of the target sample.

[0010] In some embodiments, determining the quantitative phase distribution of the refractive index of the target sample based on the preset refractive characteristic value of the target sample and the phase gradient characteristic of the refractive index specifically includes: Obtain the preset refractive feature values ​​of the target sample; The phase gradient features of the refractive light are converted into gradient field data covering the target sample; The quantitative phase distribution of the target sample refractive error is determined based on the refractive characteristics and the gradient field data.

[0011] In some embodiments, determining the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample specifically includes: Obtain the refractive index and thickness of the target sample; Obtain the wavelength parameters of the light-emitting diodes in the light-emitting diode array; The optical path difference at each position of the target sample is determined based on the quantitative phase distribution and the wavelength parameters. The refractive power distribution of the target sample is determined based on the optical path difference at each location, the refractive index, and the thickness.

[0012] Secondly, this application provides a refractive measurement system, comprising: The acquisition module is used to acquire sample images from various illumination angles during the process of illuminating the target sample with a light-emitting diode array; The processing module is used to extract angular spectral information related to the refractive properties of the target sample from each sample image, enhance the angular spectral information in each sample image, and then obtain a feature-enhanced image set. The processing module is also used to determine the refractive interference characteristics of the target sample during the refractive measurement process, and to perform gradient noise reduction on the refractive interference of the target sample based on the refractive interference characteristics and the feature enhancement image set, so as to obtain the phase gradient characteristics of the refractive error of the target sample after refractive noise reduction. The processing module is also used to determine the quantitative phase distribution of the refractive index of the target sample based on the preset refractive feature quantity of the target sample and the phase gradient feature of the refractive index. An execution module is used to determine the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample.

[0013] Thirdly, this application provides a computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described refractive measurement method.

[0014] Fourthly, this application provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described refractive measurement method.

[0015] The technical solutions provided by the embodiments disclosed in this application have the following beneficial effects: The refractive measurement system and method provided in this application first acquire sample images of the target sample at various illumination angles during the illumination of the target sample by a light-emitting diode array; extract angular spectral information related to the refractive properties of the target sample from each sample image, enhance the angular spectral information in each sample image to obtain a feature-enhanced image set; determine the refractive interference characteristics of the target sample during the refractive measurement process, perform gradient noise reduction on the refractive interference of the target sample based on the refractive interference characteristics and the feature-enhanced image set to obtain the phase gradient characteristics of the refractive error of the target sample after noise reduction; determine the quantitative phase distribution of the refractive error of the target sample based on the preset refractive feature quantity of the target sample and the phase gradient characteristics of the refractive error; and determine the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample.

[0016] Therefore, this application, during refractive measurement, acquires comprehensive optical information of the target sample by collecting sample images from various illumination angles, avoiding interference and misjudgment caused by missing information from a single angle. Extracting and enhancing angular spectral information allows focusing on core features directly related to refractive properties, weakening irrelevant noise interference, and improving feature recognition. Targeted determination of refractive interference features and implementation of gradient denoising accurately remove interference and noise while preserving the integrity and accuracy of refractive phase gradient features to the greatest extent, solving the problem of traditional denoising easily losing effective information. Combining preset refractive feature quantities to determine quantitative phase distribution, and further calibrating the data through feature matching, reduces the impact of residual interference. Finally, combining refractive index and thickness to calculate the refractive power distribution, achieving accurate quantification of refractive power based on precise preprocessed data. Using the above scheme, the influence of optical interference and noise on refractive measurement under multi-angle illumination can be overcome. Attached Figure Description

[0017] Figure 1 This is an exemplary flowchart of a refractive measurement method according to some embodiments of this application; Figure 2 This is an exemplary flowchart illustrating the determination of angular spectrum information according to some embodiments of this application; Figure 3This is an exemplary flowchart illustrating the determination of refractive interference characteristics according to some embodiments of this application; Figure 4 This is a schematic diagram of the refractive measurement system according to some embodiments of this application; Figure 5 This is a schematic diagram of the structure of a computer device for implementing a refractive measurement method according to some embodiments of this application. Detailed Implementation

[0018] To better understand the technical solution of this application, the technical solution of this application will be described in detail below with reference to the accompanying drawings and specific embodiments.

[0019] refer to Figure 1 The figure is an exemplary flowchart of a refractive measurement method according to some embodiments of this application, which mainly includes the following steps: In some embodiments, multi-angle illumination of the target sample of the refractive error under test using an array of light-emitting diodes can be achieved by sequentially lighting the light-emitting diodes at corresponding positions in the array according to a preset angle sequence, and setting the illumination duration of each angle to match the exposure time of image acquisition, so as to achieve multi-angle illumination of the target sample.

[0020] In step 101, sample images of each illumination angle are acquired during the process of illuminating the target sample with the light-emitting diode array.

[0021] It should be noted that the sample images in this application include optical images of the target sample acquired by the light-emitting diode array under different illumination angles. The images contain details of the refractive structure of the target sample and optical information on the interaction between light and the sample. The sample images reflect the light reflection, refraction and scattering characteristics of the target sample under different illumination angles, highlighting the detailed features of the refractive-related structural edges and grayscale distribution. The target sample can be a human eye or an artificial lens.

[0022] In specific implementation, when the corresponding diode is triggered to light up according to the preset lighting angle sequence, a collection signal is sent to the camera simultaneously. The camera takes 1-3 images at each lighting angle to ensure the reliability of the collection. After the images are taken, the images are immediately preliminarily screened to remove blurry, overexposed or underexposed images caused by accidental interference, so as to obtain sample images containing each lighting angle. Other methods can also be used for collection in other embodiments, which will not be elaborated here.

[0023] In step 102, angular spectral information related to the refractive properties of the target sample is extracted from each sample image, and the angular spectral information in each sample image is enhanced to obtain a feature-enhanced image set.

[0024] In some embodiments, reference Figure 2 As shown, this figure is an exemplary flowchart for determining angular spectral information in some embodiments of this application. In this embodiment, the extraction of angular spectral information related to the refractive properties of the target sample from each sample image can be achieved by the following steps: In step 1021, each sample image is denoised to obtain each denoised sample image; In step 1022, the refractive properties of the target sample are obtained; In step 1023, angular spectral information related to the refractive properties of the target sample is extracted from each denoised sample image.

[0025] In specific implementation, each sample image is denoised to obtain a denoised sample image. This can be achieved by using a 3×3 or 5×5 window Gaussian filter, setting the filter kernel variance according to the statistical characteristics of image noise, replacing the center pixel by calculating the weighted average of pixels within the window to efficiently suppress random Gaussian noise, and combining this with a 3×3 window median filter to traverse the image and replace the center pixel with the median of pixels within the window to specifically eliminate salt-and-pepper noise, while preserving refractive-related edge details to the greatest extent, ultimately obtaining a denoised sample image with a high signal-to-noise ratio. Other denoising methods can also be used in other embodiments, which are not limited here.

[0026] It should be noted that the refractive properties in this application reflect the target sample's ability to deflect, focus, or diverge incident light. The refractive properties represent the inherent optical characteristics of the sample itself related to light propagation. These characteristics directly determine the propagation trajectory of light after passing through the sample, the image clarity, and the refractive correction requirements. The refractive properties include the sample's refractive index, physical thickness, surface and internal curvature, refractive sensitive frequency range, angular response law, and specific refractive states such as myopia, hyperopia, and astigmatism unique to biological samples like the human eye. At the same time, it also covers key information on the differences in optical properties between the sample and the surrounding medium.

[0027] In addition, in specific implementation, extracting the angular spectrum information related to the refractive properties of the target sample from each denoised sample image can be achieved in the following way: Perform a two-dimensional fast Fourier transform on each denoised sample image to convert the image from the spatial domain to the frequency domain to obtain a complete spectrum. Based on the known refractive properties, set precise frequency domain screening conditions. That is, first, match with the standard refractive property database through preliminary calibration experiments to determine the specific numerical range of the refractive sensitive frequency of the target sample. For example, the refractive sensitive frequency of the human cornea usually corresponds to the range of 0.05-0.2 cycles / pixel. Specifically, it needs to be corrected in combination with the sample refractive index, thickness and illumination system parameters. At the same time, the typical frequency distribution of ambient light interference and optical system aberrations is statistically analyzed as the exclusion range. The screening condition is to retain the signal within the refractive sensitive frequency range and eliminate the signal in the high and low frequency interference range. A customized Gaussian bandpass filter is used, and the filter center frequency is aligned with the refractive sensitive frequency. The bandwidth is adjusted according to the frequency response range of the refractive properties. The filtering curve uses a Gaussian function weighting to achieve a smooth transition. The spectrum is selectively screened. First, the spectrum is multiplied with the frequency response function of the Gaussian bandpass filter at frequency points to preserve and slightly enhance the signal in the refractive sensitive frequency range, while the signal in the high and low frequency interference range gradually attenuates according to the Gaussian curve. Then, by setting an amplitude threshold of 1 / 10 of the average amplitude of the signal in the refractive sensitive range, the frequency components with amplitudes below the threshold after the product operation are directly set to zero to further eliminate residual weak interference. Finally, the amplitude of the processed spectrum is normalized to ensure that the amplitude of the spectrum after screening of different images is consistent, effectively eliminating irrelevant frequency components such as ambient light interference and optical system aberrations. Finally, the angular spectrum information directly related to the refractive properties of the target sample is extracted from the screened target spectrum region. Other extraction methods can be used in other embodiments, which are not limited here.

[0028] It should be noted that the angular spectrum information in this application represents a set of specific frequency components in the sample image that are directly related to the refractive properties of the target sample. It can be used to express the refraction, focusing or divergence characteristics of incident light at different angles in the frequency domain. The angular spectrum information includes the frequency position corresponding to the refractive sensitive frequency range, the amplitude, phase distribution, frequency bandwidth of each frequency component, and the distribution pattern of different frequency components.

[0029] In some embodiments, enhancing the angular spectral information in each sample image to obtain a feature-enhanced image set can be achieved through the following steps: The angular spectral information in each sample image is enhanced to obtain the enhanced angular spectral information of each sample image; The feature-enhanced image set is determined based on the enhanced angular spectrum information in each sample image.

[0030] In practice, the angular spectrum information in each sample image is enhanced. This enhanced angular spectrum information can be achieved as follows: First, the extracted angular spectrum information is enhanced in stages. The first step uses a frequency domain Gaussian low-pass filter. Based on the statistical characteristics of noise in the angular spectrum, the filter kernel size is set to 3×3 or 5×5; a larger kernel is used when the noise intensity is high. The variance is estimated using the noise standard deviation of the sample images, typically between 0.8 and 1.2. High-frequency random noise is filtered out while retaining refractive-sensitive frequency components by calculating the weighted average of each frequency point and its neighborhood in the frequency domain. The second step applies the Laplacian sharpening operator to enhance details. The Laplacian operator is transformed to the frequency domain, and its frequency response is ∇. 2 F(fx,fy)=-4π 2 (fx 2 +fy 2 F(fx,fy), where F is the angular spectrum and fx and fy are frequency coordinates, is multiplied point by point with the angular spectrum data to enhance high-frequency details related to refractive characteristics, such as frequency components corresponding to changes in sample edge curvature and differences in refractive index distribution; the third step is to perform spectral amplitude histogram equalization, statistically analyze the histogram distribution of angular spectrum amplitudes, calculate the cumulative distribution function, and map the original amplitudes to a uniformly distributed grayscale range to improve the contrast of weak refractive signals and avoid feature submersion caused by excessive amplitude differences. Finally, the enhanced angular spectrum information in each sample image is obtained. Other methods can be used in other embodiments, which are not limited here.

[0031] In addition, in specific implementation, the feature enhancement image set can be determined based on the enhanced angular spectrum information in each sample image in the following way: perform a two-dimensional inverse fast Fourier transform on the enhanced angular spectrum information to convert the frequency domain signal back to the spatial domain, and obtain the feature enhancement image of a single sample; process the angular spectrum information of all sample images in sequence according to the above process, and arrange the inverse transformed spatial domain images in the order of the original illumination angle to finally form the feature enhancement image set. Other methods can also be used to determine the feature enhancement image set in other embodiments, which are not limited here.

[0032] It should be noted that the enhanced angular spectrum information in this application represents optimized data after the frequency features directly related to the refractive properties of the target sample are enhanced and irrelevant interference frequencies are suppressed, which makes the frequency response corresponding to the refractive properties of the sample clearer, and improves the identification of weak refractive signals and the stability of strong refractive signals; the feature-enhanced image set represents the image set in the spatial domain after the refractive features of the sample are highlighted, which makes the spatial distribution of the refractive properties of the sample under different illumination angles more intuitive and clear, and significantly reduces noise interference.

[0033] In step 103, the refractive interference characteristics of the target sample during the refractive measurement process are determined. Based on the refractive interference characteristics and the feature enhancement image set, gradient denoising is performed on the refractive interference of the target sample to obtain the phase gradient characteristics of the refractive error of the target sample after refractive denoising.

[0034] In some embodiments, reference Figure 3 As shown, this figure is an exemplary flowchart of determining refractive interference characteristics in some embodiments of this application. In this embodiment, determining the refractive interference characteristics of the target sample during the refractive measurement process can be achieved by the following steps: In step 1031, various interference sources of the target sample during the refractive measurement process are obtained; In step 1032, the interference characteristics of each interference source are determined; In step 1033, the refractive interference characteristics of the target sample during the refractive measurement process are determined based on all the interference characteristics.

[0035] It should be noted that the interference sources of the target sample in the refractive measurement process in this application include five categories: optical system, target sample itself, measurement environment, illumination source and image sensor. Specifically, these include inherent aberrations of the optical system such as spherical aberration, chromatic aberration, and distortion, as well as optical path interference caused by lens reflection and stray light from the lens barrel; scattering and absorption of the target sample itself, and dust or stains adhering to its surface; ambient stray light in the measurement environment and optical component parameter drift caused by temperature and humidity changes; brightness fluctuations, emission angle drift, spectral shift, and uneven brightness between LEDs of the illumination source; and dark current noise, readout noise, inconsistent pixel response, and thermal noise of the image sensor.

[0036] In practical implementation, determining the interference characteristics of each interference source can be achieved in the following way: For each interference source, its interference characteristics are determined individually. For optical system aberrations, the Zernike polynomial fitting method is used. First, the point spread function of the point source through the system is collected. A Fourier transform is performed on the point spread function to obtain the optical transfer function. Then, the phase component of the optical transfer function is fitted using Zernike polynomials to extract characteristic parameters such as the order and coefficients of the aberrations. For sample scattering / absorption, the diffuse reflectance and transmittance of the sample are measured using an integrating sphere. The scattering coefficient and absorption coefficient are calculated, and the spatial distribution characteristics of the scattering are analyzed in conjunction with dark-field imaging. For ambient light interference... Blank images were acquired in both darkroom and natural light environments. The intensity distribution and frequency characteristics of ambient light were determined by grayscale variance analysis. For sensor noise, 100 dark-field images were acquired with the light source blocked. The mean and standard deviation of pixel grayscale were calculated, and the frequency distribution of noise was analyzed by power spectral density. For unstable light sources, the light intensity was monitored in real time using photodiodes. The light intensity fluctuation coefficient (standard deviation / mean) was calculated to reflect brightness stability. The change in the emission angle was measured using a laser interferometer to reflect the angle drift characteristics. Finally, the interference characteristics of each interference source were obtained. Other methods can be used to determine the interference characteristics in other embodiments, which are not limited here.

[0037] In addition, in specific implementation, the determination of the refractive interference characteristics of the target sample in the refractive measurement process based on all interference characteristics can be achieved in the following way: integrate all interference characteristics, first quantify each interference characteristic into a unified index by the minimum-maximum normalization method, that is, map it to the [0,1] interval, then analyze the correlation between each index and the refractive measurement error through multiple linear regression, and assign weights according to the magnitude of the correlation, that is, take the absolute value of the correlation corresponding value of all interference characteristics, and then perform normalization processing, divide the correlation corresponding value after each absolute value by the sum of the absolute values ​​of all correlation corresponding values, and the ratio obtained is the weight of the interference characteristic. Calculate the weighted sum of each interference characteristic and its corresponding weight, and use the result of the weighted sum calculation as the refractive interference characteristics of the target sample in the measurement process.

[0038] It should be noted that the interference features in this application represent the potential influence mode and intensity of the interference source on the refractive measurement signal; the refractive interference features represent the overall interference state under the synergistic effect of various interferences during the refractive measurement process, and can be used to analyze the comprehensive influence trend and the dominant direction of the core interference on the accuracy of refractive parameter measurement.

[0039] In some embodiments, the refractive interference of the target sample is subjected to gradient denoising based on the refractive interference features and the feature-enhanced image set to obtain the phase gradient features of the refractive error of the target sample after refractive denoising. This can be achieved by the following steps: Select one feature-enhanced image from the feature-enhanced image set as the selected feature-enhanced image, and determine the initial phase gradient map of the selected feature-enhanced image; Based on the refractive interference characteristics, determine the effective gradient region and interference gradient region of the refractive field in the initial phase gradient map; The interference gradient region is subjected to interference suppression to obtain the interference gradient region after interference suppression; The effective gradient region is denoised to obtain the denoised effective gradient region. The phase gradient of the refractive image after refractive denoising is determined based on the interference gradient region after interference suppression and the effective gradient region after denoising. Continue to determine the phase gradient of the refractive image after refractive noise reduction in the image enhanced by remaining features; All phase gradients are used as the phase gradient features of the refractive image after refractive noise reduction of the target sample.

[0040] In specific implementation, determining the initial phase gradient map of the selected feature enhancement image can be achieved as follows: The gray-level gradients in the x and y directions of the image are calculated using a 3×3 window Sobel operator. The horizontal and vertical gradient magnitudes are obtained through convolution operations, and then the combined gradient magnitude and gradient direction are calculated. Specifically, two 3×3 convolution kernels of the Sobel operator are determined: the kernel Gx used to calculate the horizontal gradient is [[-1,0,1],[-2,0,2],[-1,0,1]], and the kernel Gy used to calculate the vertical gradient is [[...]. -1,-2,-1],[0,0,0],[1,2,1]]; If the selected feature enhancement image is a color image, it needs to be converted to a single-channel grayscale image first using the weighted average method to ensure the uniformity of gradient calculation; then, the grayscale image is traversed pixel by pixel, and a 3×3 neighborhood is taken with each pixel as the center. The grayscale value of the pixel in the neighborhood is multiplied by the corresponding elements of Gx and Gy respectively, and then summed to obtain the horizontal gradient value Gx and the vertical gradient value Gy of the pixel. Edge pixels need to be filled with the neighborhood using the zero-padding method to avoid the loss of boundary information; then, according to the formula gradient magnitude = The gradient intensity of each pixel is calculated to reflect the significance of the change in refractive characteristics at that location. At the same time, the gradient angle is calculated using the formula gradient direction = arctan(Gy / Gx), which ranges from -π to π and corresponds to the direction of light deflection. Finally, the gradient magnitude and direction of all pixels are combined to form a two-dimensional matrix with intensity-direction as the core, i.e., the initial phase gradient map. In other embodiments, other methods can be used to determine the gradient, which are not limited here.

[0041] In addition, in specific implementation, the determination of the effective gradient region and interference gradient region of the refractive error in the initial phase gradient map based on the refractive interference characteristics can be achieved in the following way: Subsequently, based on the refractive interference characteristics, the gradient amplitude is filtered by adaptive threshold segmentation. The threshold is set based on the statistical value of the maximum amplitude of the interference gradient, for example, taking 1.2 times the maximum influence amplitude of all interference sources in the interference characteristics. At the same time, a secondary judgment is made based on the consistency of the direction of the effective refractive gradient. For example, the consistency of direction can be that the effective gradient direction of the human eye sample is mostly parallel to the tangent direction of the corneal curvature. The region with gradient amplitude below the threshold and discrete direction is defined as the interference gradient region, and the region with gradient amplitude above the threshold and concentrated direction is defined as the effective gradient region. Other methods can also be used to determine this in other embodiments, which are not limited here.

[0042] In addition, in specific implementation, interference suppression is performed on the interference gradient region. The interference-suppressed interference gradient region can be obtained in the following way: weighted nonlocal mean filtering is used to suppress interference in the interference gradient region. The size of the filter kernel is dynamically adjusted according to the interference weight of the region. A 5×5 window is used when the interference weight is high, and a 3×3 window is used when the weight is low. The weight is assigned by calculating the gray-level similarity between the target pixel and the neighboring pixels. The higher the similarity, the greater the weight. That is, the neighborhood window is determined with the target pixel as the center. Each neighboring pixel in the window is traversed, and the gray-level similarity between the target pixel and the neighboring pixels is calculated. The absolute value of the gray-level difference is used as the similarity metric, that is, similarity S=1 / (1+|I_t-I_n|), where I_t is the gray value of the target pixel and I_n is the gray value of the neighboring pixel. The smaller |I_t-I_n| is, the closer S is to 1, and the higher the similarity is. Then, a Gaussian smoothing factor is introduced to optimize the weight allocation, and the weight w=S×exp(-d 2 / (2σ 2 )), where d is the spatial distance between the target pixel and its neighboring pixels, and σ is the spatial distance attenuation coefficient, usually taken as 1.0~2.0, to further strengthen the influence of spatial proximity on the weights; finally, the weights of all neighboring pixels in the window are normalized to ensure that the sum of all weights is 1, so as to realize the allocation logic that the neighboring pixels with higher gray-level similarity and closer spatial distance have a larger weight proportion. The original pixel value is replaced with the weighted average result to strongly weaken the interference component, and finally the interference gradient region after interference suppression is obtained. Other methods can be used to suppress interference in other embodiments, which are not limited here.

[0043] In addition, in specific implementation, denoising the effective gradient region can be achieved by the following method: Denoising the effective gradient region using bilateral filtering. First, set the spatial domain standard deviation to 1.5 to control the influence range of the spatial neighborhood, and set the grayscale domain standard deviation to 20 to control the influence of grayscale similarity. Traverse each target pixel in the region and construct a 5×5 neighborhood window. For each neighboring pixel within the window, apply the Gaussian function W_s(d)=exp(-d 2 / (2×1.5 2 ), where d is the Euclidean distance between the target pixel and its neighboring pixels. Spatial weights are calculated using the Gaussian function W_r(ΔI)=exp(-ΔI). 2 / (2×20 2 In this process, ΔI represents the grayscale difference between the two pixels. The grayscale weights are calculated, and the two weights are multiplied to obtain the total weight. Only spatially adjacent pixels with similar grayscale values ​​receive high weights, while noisy pixels that are spatially distant or have abrupt grayscale changes receive low weights. Subsequently, the grayscale values ​​of all neighboring pixels within the window are weighted and averaged according to the total weight. The filtered pixel value is calculated using the formula: Σ(total weight × neighboring pixel grayscale value) / Σ total weight. The calculated result is used to replace the original grayscale value of the target pixel. Finally, while smoothing out isolated small noises, the fine gradient details related to refractive power are fully preserved to obtain the effective gradient region after denoising. Other methods can also be used for denoising in other embodiments, which are not limited here.

[0044] In addition, in specific implementation, determining the phase gradient of the selected feature-enhanced image after refractive noise reduction based on the interference gradient region after interference suppression and the effective gradient region after denoising can be achieved in the following way: The boundary line between the two regions and the transition region near the boundary are located by using a region mask. The mask value of the effective gradient region is set to 1, and the mask value of the interference gradient region is set to 0. The transition region is usually a range of 1-2 pixels on both sides of the boundary line to avoid abrupt transitions. Then, for each pixel in the transition region, its spatial distance to the effective gradient region and the interference gradient region is calculated as the basis for weight allocation. Let the distance from a pixel in the transition region to the boundary of the effective gradient region be d1, and the distance to the boundary of the interference gradient region be d2. The total distance d = d1 + d2, then the effective region weight w of that pixel is... 1 = d2 / d, the weight of the interference region w2 = d1 / d. The closer to a certain region, the greater the weight of the corresponding region, to achieve a natural transition. Then, the fused gray value of a pixel in the transition region is calculated. The fused value = (gray value of the pixel corresponding to the effective region after denoising × w1) + (gray value of the pixel corresponding to the interference region after interference suppression × w2). For pixels inside the effective gradient region, their denoised gray value is directly retained. For pixels inside the interference gradient region, their interference suppressed gray value is directly retained. Finally, the entire image is traversed pixel by pixel according to the above rules to complete the integration of all regions. Finally, the phase gradient of the selected feature-enhanced image after refractive noise reduction is obtained with smooth boundaries, no gray-level abrupt changes, and continuous gradient information. Other methods can be used to determine this in other embodiments, which are not limited here.

[0045] It should be noted that the initial phase gradient map in this application represents the original gradient distribution state of all refractive-related changes and interferences in the image; the effective gradient region represents the core gradient range directly related to the refractive properties of the target sample, reflecting the real refractive characteristics such as corneal curvature changes and refractive index distribution differences; the interference gradient region represents the gradient range affected by interference sources, reflecting the destructive effect of various interferences on the gradient signal; the phase gradient represents the changing trend of refractive characteristics at this location, reflecting the local light deflection ability and refractive state of the target sample; the phase gradient feature represents the overall gradient information of the target sample under full illumination angle, reflecting the spatial distribution law and angular response characteristics of the overall refractive characteristics of the sample.

[0046] In step 104, the quantitative phase distribution of the refractive power of the target sample is determined based on the preset refractive feature quantity of the target sample and the phase gradient feature of the refractive power.

[0047] In some embodiments, determining the quantitative phase distribution of the refractive index of a target sample based on preset refractive characteristic values ​​and the phase gradient characteristics of the refractive index can be achieved through the following steps: Obtain the preset refractive feature values ​​of the target sample; The phase gradient features of the refractive light are converted into gradient field data covering the target sample; The quantitative phase distribution of the target sample refractive error is determined based on the refractive characteristics and the gradient field data.

[0048] It should be noted that the refractive characteristics in this application include the geometric parameters of the sample, such as the thickness distribution range and surface curvature radius; optical parameters, such as the mean refractive index of the sample material and the difference in refractive index with the surrounding medium; and measurement system parameters, such as the wavelength of the illumination source and the pixel size and magnification of the imaging system. The refractive characteristics represent the core quantitative information of the sample's own physical structure, optical properties, and measurement environment. The refractive characteristics reflect the sample's inherent ability to refract, focus, or diverge light.

[0049] In specific implementation, the conversion of the phase gradient features of the refraction into gradient field data covering the target sample can be achieved in the following way: Since the phase gradient features are discrete gradient information under different illumination angles, it is necessary to first align the gradient maps of all angles to the same spatial coordinate system through image registration technology. With the sample center as the origin, a two-dimensional pixel coordinate system of x and y is established. Then, for each spatial pixel (x, y), the gradient magnitude and direction under all illumination angles are integrated to construct three-dimensional gradient field data. The dimension is pixel x coordinate × pixel y coordinate × illumination angle. In other embodiments, other methods can also be used to determine the dimension, which is not limited here.

[0050] In addition, in specific implementation, the quantitative phase distribution of the target sample refractive error can be determined based on the refractive characteristics and the gradient field data in the following manner: The quantitative phase distribution is determined using a least-squares phase recovery algorithm. First, based on the refractive index difference Δn and illumination wavelength λ in the refractive characteristics, a physical correlation model between phase φ and gradient G is established (φ = (2π / λ) × Δn × d, where d is the sample thickness, and G = ∇φ, i.e., the gradient is the spatial derivative of the phase). Then, the gradient field data is integrated using a step-by-step integration method. First, the gradient of each y-coordinate is integrated along the x-axis to obtain the preliminary phase value for each row, and then the gradient is integrated along the y-axis. The correction involves incorporating gradient weights for each illumination angle, with the center angle weight set to 1.0 and the edge angle weights decreasing to 0.5 according to a cosine law, balancing the imaging contributions from different angles. Finally, a Tikhonov regularization term is introduced, with the regularization parameter determined through cross-validation, typically ranging from 0.01 to 0.1, to suppress noise amplification during integration. Through iterative optimization, the calculated phase distribution is optimized to minimize the error between the gradient and the measured gradient field and ensure a smooth phase distribution. Ultimately, a quantitative phase distribution covering the entire target sample and reflecting the spatial variation of its refractive properties is obtained, in radians. Other methods can be used to determine this phase distribution in other embodiments, which are not limited here.

[0051] It should be noted that the gradient field data in this application reflects the changing trend of the refractive properties of the target sample at each position and the gradient response law corresponding to incident light at different angles; the quantitative phase distribution represents the distribution of the absolute phase value at each spatial position of the target sample, reflecting the amount of phase delay caused by the difference in refractive properties after the light passes through the target sample, and is directly related to the actual refractive power and spatial distribution law of the sample.

[0052] In step 105, the refractive power distribution of the target sample is determined based on the quantitative phase distribution and the refractive index and thickness of the target sample.

[0053] In some embodiments, determining the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample can be achieved using the following steps: Obtain the refractive index and thickness of the target sample; Obtain the wavelength parameters of the light-emitting diodes in the light-emitting diode array; The optical path difference at each position of the target sample is determined based on the quantitative phase distribution and the wavelength parameters. The refractive power distribution of the target sample is determined based on the optical path difference at each location, the refractive index, and the thickness.

[0054] In specific implementation, the optical path difference at each position of the target sample can be determined according to the quantitative phase distribution and the wavelength parameter in the following way: calculate the optical path difference OPD(x,y) at each position. Based on the physical relationship between the quantitative phase distribution and the optical path difference, φ=2π・OPD / λ, where φ is the phase value of the quantitative phase distribution in radians and λ is the wavelength parameter, and the values ​​are substituted pixel by pixel to obtain OPD(x,y)=φ(x,y)・λ / (2π), and the unit is converted to meters. The optical path difference reflects the optical path delay between the sample and the air.

[0055] Furthermore, in practical implementation, the refractive power distribution of the target sample can be determined based on the optical path difference at each location, the refractive index, and the thickness in the following manner: Under the paraxial approximation, combined with the definition of optical path difference OPD=(n-n0)t, the refractive power D(x,y) (unit D=1 / m) is related to OPD and radial distance ρ(x,y)= Where x and y are physical coordinates, transformed from pixel size s: physical x = pixel x・s × 10 -6 The relationship between m) is D(x,y)=2・OPD(x,y) / (ρ(x,y)). 2 The refractive power distribution of the target sample is obtained by calculating ρ(x,y) pixel by pixel and substituting it into the formula. If the target sample is aspherical, the calculation can be optimized by fitting the OPD with Zernike polynomials to ensure that the result accurately reflects the refractive power of each position of the sample.

[0056] It should be noted that the refractive power distribution in this application represents the numerical distribution of refractive power at various spatial locations of the target sample, reflecting the ability of different regions of the sample to focus or diverge light. For example, the refractive power distribution of the cornea can reflect the refractive difference between the central and peripheral regions, and the distribution of engineering lenses can reflect the uniformity of optical design. At the same time, it can also intuitively show whether there are refractive abnormalities in the sample, such as the asymmetrical distribution corresponding to astigmatism, and local excessive / low refractive power.

[0057] In another aspect, in some embodiments, this application provides a refractive measurement system, with reference to... Figure 4 The figure is a schematic diagram of the structure of a refractive measurement system according to some embodiments of this application. The refractive measurement system 400 includes: an acquisition module 401, a processing module 402, and an execution module 403, which are described below: Acquisition module 401, in this application, is mainly used to acquire sample images from various illumination angles during the process of illuminating the target sample with a light-emitting diode array; Processing module 402, in this application, is used to extract angular spectral information related to the refractive properties of the target sample from each sample image, enhance the angular spectral information in each sample image, and then obtain a feature-enhanced image set. It should be noted that the processing module 402 in this application is also used to determine the refractive interference characteristics of the target sample during the refractive measurement process, and to perform gradient denoising on the refractive interference of the target sample according to the refractive interference characteristics and the feature enhancement image set, so as to obtain the phase gradient characteristics of the refractive error of the target sample after refractive denoising. In addition, it should be noted that the processing module 402 in this application is also used to determine the quantitative phase distribution of the refractive power of the target sample based on the preset refractive feature quantity of the target sample and the phase gradient feature of the refractive power. The execution module 403 in this application is mainly used to determine the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample.

[0058] In addition, this application also provides a computer device, the computer device including a memory and a processor, the memory storing code, and the processor being configured to acquire the code and execute the above-described refractive measurement method.

[0059] In some embodiments, reference Figure 5 The figure is a schematic diagram of the structure of a computer device for implementing a refractive measurement method according to some embodiments of this application. The refractive measurement method in the above embodiments can be achieved through... Figure 5 The computer device shown is used to implement this, and the computer device 500 includes at least one processor 501, a communication bus 502, a memory 503, and at least one communication interface 504.

[0060] Processor 501 can be a general-purpose central processing unit (CPU) or an application-specific integrated circuit (ASIC).

[0061] The communication bus 502 can be used to transmit information between the aforementioned components.

[0062] Memory 503 may be a read-only memory (ROM) or other type of static storage device capable of storing static information and instructions, random access memory (RAM) or other type of dynamic storage device capable of storing information and instructions, or electrically erasable programmable read-only memory (EEPROM), compact disc read-only memory (CDROM) or other optical disc storage, optical disc storage (including compressed optical discs, laser discs, optical discs, digital versatile optical discs, Blu-ray discs, etc.), magnetic disks or other magnetic storage devices, or any other medium capable of carrying or storing desired program code in the form of instructions or data structures and accessible by a computer, but not limited thereto. Memory 503 may exist independently and be connected to processor 501 via communication bus 502. Memory 503 may also be integrated with processor 501.

[0063] The memory 503 stores program code for executing the scheme of this application, and its execution is controlled by the processor 501. The processor 501 executes the program code stored in the memory 503. The program code may include one or more software modules. The method used in the above embodiments can be implemented by the processor 501 and one or more software modules in the program code in the memory 503.

[0064] Communication interface 504 uses any transceiver-like device to communicate with other devices or communication networks, such as Ethernet, radio access network (RAN), wireless local area networks (WLAN), etc.

[0065] In a specific implementation, as one example, a computer device may include multiple processors, each of which may be a single-core (single CPU) processor or a multi-core (multi CPU) processor. Here, a processor may refer to one or more devices, circuits, and / or processing cores used to process data (e.g., computer program instructions).

[0066] The aforementioned computer device can be a general-purpose computer device or a special-purpose computer device. In specific implementations, the computer device can be a desktop computer, a portable computer, a network server, a handheld digital assistant (PDA), a mobile phone, a tablet computer, a wireless terminal device, a communication device, or an embedded device. This application does not limit the type of computer device.

[0067] In addition, this application also provides a computer-readable storage medium storing a computer program that, when executed by a processor, implements the above-described refractive measurement method.

[0068] Although preferred embodiments of this application have been described, those skilled in the art, upon learning the basic inventive concept, can make other changes and modifications to these embodiments. Therefore, the appended claims are intended to be interpreted as including the preferred embodiments as well as all changes and modifications falling within the scope of this application.

[0069] Obviously, those skilled in the art can make various modifications and variations to this application without departing from the spirit and scope of this application. Therefore, if such modifications and variations fall within the scope of the claims of this application and their equivalents, this application also intends to include such modifications and variations.

Claims

1. A method for measuring refractive power, wherein, A method for illuminating a target sample of refractive power under test from multiple angles using a light-emitting diode array is characterized by the following steps: Acquire sample images from various illumination angles during the process of illuminating the target sample with a light-emitting diode array; Angular spectral information related to the refractive properties of the target sample is extracted from each sample image, and the angular spectral information in each sample image is enhanced to obtain a feature-enhanced image set. Determine the refractive interference characteristics of the target sample during the refractive measurement process, and perform gradient denoising on the refractive interference of the target sample based on the refractive interference characteristics and the feature-enhanced image set to obtain the phase gradient characteristics of the refractive error of the target sample after refractive denoising. The quantitative phase distribution of the refractive index of the target sample is determined based on the preset refractive characteristic quantity of the target sample and the phase gradient characteristic of the refractive index. The refractive power distribution of the target sample is determined based on the quantitative phase distribution and the refractive index and thickness of the target sample.

2. The method as described in claim 1, characterized in that, Extracting spectral information related to the refractive properties of the target sample from each sample image specifically includes: Denoise the individual sample images to obtain the denoised sample images; Obtain the refractive properties of the target sample; Angular spectral information related to the refractive properties of the target sample is extracted from each denoised sample image.

3. The method as described in claim 1, characterized in that, The angular spectral information in each sample image is enhanced to obtain the feature-enhanced image set, which specifically includes: The angular spectral information in each sample image is enhanced to obtain the enhanced angular spectral information of each sample image; The feature-enhanced image set is determined based on the enhanced angular spectrum information in each sample image.

4. The method as described in claim 1, characterized in that, Determining the refractive interference characteristics of the target sample during refractive measurement specifically includes: Identify all interference sources in the target sample during refractive measurement. Determine the interference characteristics of each interference source; The refractive interference characteristics of the target sample during the refractive measurement process are determined based on all interference characteristics.

5. The method as described in claim 1, characterized in that, Based on the refractive interference features and the feature-enhanced image set, gradient denoising is performed on the refractive interference of the target sample to obtain the phase gradient features of the refractive error after denoising of the target sample. Specifically, this includes: Select one feature-enhanced image from the feature-enhanced image set as the selected feature-enhanced image, and determine the initial phase gradient map of the selected feature-enhanced image; Based on the refractive interference characteristics, determine the effective gradient region and interference gradient region of the refractive field in the initial phase gradient map; The interference gradient region is subjected to interference suppression to obtain the interference gradient region after interference suppression; The effective gradient region is denoised to obtain the denoised effective gradient region. The phase gradient of the refractive image after refractive denoising is determined based on the interference gradient region after interference suppression and the effective gradient region after denoising. Continue to determine the phase gradient of the refractive image after refractive noise reduction in the image enhanced by remaining features; All phase gradients are used as the phase gradient features of the refractive image after refractive noise reduction of the target sample.

6. The method as described in claim 1, characterized in that, Determining the quantitative phase distribution of the target sample's refractive index based on the preset refractive characteristic values ​​and the phase gradient characteristics of the refractive index specifically includes: Obtain the preset refractive feature values ​​of the target sample; The phase gradient features of the refractive light are converted into gradient field data covering the target sample; The quantitative phase distribution of the target sample refractive error is determined based on the refractive characteristics and the gradient field data.

7. The method as described in claim 1, characterized in that, Determining the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample specifically includes: Obtain the refractive index and thickness of the target sample; Obtain the wavelength parameters of the light-emitting diodes in the light-emitting diode array; The optical path difference at each position of the target sample is determined based on the quantitative phase distribution and the wavelength parameters. The refractive power distribution of the target sample is determined based on the optical path difference at each location, the refractive index, and the thickness.

8. A refractive measurement system, characterized in that, include: The acquisition module is used to acquire sample images from various illumination angles during the process of illuminating the target sample with a light-emitting diode array; The processing module is used to extract angular spectral information related to the refractive properties of the target sample from each sample image, enhance the angular spectral information in each sample image, and then obtain a feature-enhanced image set. The processing module is also used to determine the refractive interference characteristics of the target sample during the refractive measurement process, and to perform gradient noise reduction on the refractive interference of the target sample based on the refractive interference characteristics and the feature enhancement image set, so as to obtain the phase gradient characteristics of the refractive error of the target sample after refractive noise reduction. The processing module is also used to determine the quantitative phase distribution of the refractive power of the target sample based on the preset refractive feature quantity of the target sample and the phase gradient feature of the refractive power; An execution module is used to determine the refractive power distribution of the target sample based on the quantitative phase distribution and the refractive index and thickness of the target sample.

9. A computer device, characterized in that, The computer device includes a memory and a processor, the memory storing code, and the processor being configured to retrieve the code and execute the refractive measurement method as described in any one of claims 1 to 7.

10. A computer-readable storage medium storing a computer program, characterized in that, When the computer program is executed by the processor, it implements the refractive measurement method as described in any one of claims 1 to 7.

Images