CN118009934A - A method and system for detecting surface roughness of an optical lens - Google Patents

Abstract

The invention discloses a method and a system for detecting the surface roughness of an optical lens, wherein the method comprises the steps of obtaining a first speckle image and a second speckle image formed by imaging after laser with different wavelengths is irradiated on the surface of the optical lens to be detected; the method comprises the steps of carrying out image enhancement on a first speckle image and a second speckle image, carrying out texture feature and gray average feature extraction on the enhanced first speckle image, carrying out texture feature, gray standard deviation and gray root mean square feature extraction on the enhanced second speckle image, respectively inputting the extracted features of the first speckle image and the extracted features of the second speckle image into a support vector machine model to obtain a first detection value and a second detection value, and taking the average value of the first detection value and the second detection value as the surface roughness of an optical lens. According to the invention, the speckle images formed by different wavelengths are respectively extracted with different features, so that the detection result is prevented from being interfered by the redundancy of the features during the model detection, the detection efficiency is improved, and the detection precision of the surface roughness is improved.

Description

A method and system for detecting surface roughness of an optical lens

Technical Field

The present invention relates to the field of optical measurement technology, and in particular to a method and system for detecting the surface roughness of an optical lens.

Background technique

Optical components usually require precision machining, and the roughness of their surface has an important influence on the performance of optical components. Common surface roughness detection methods include light sectioning microscope measurement and traditional interferometry. In light sectioning microscope measurement, the light emitted by the light source passes through a condenser and passes through a slit to form a ribbon beam. The light beam then passes through the objective lens and shoots toward the workpiece at a 45-degree angle, presenting a zigzag light band on the uneven surface, and then reflects at a 45-degree angle through the objective lens to reach the graticule. The zigzag bright band seen from the eyepiece has two boundaries. The degree of tortuosity of the light band image boundary indicates the peak and valley height of the image, which is used to measure the surface roughness. Interferometry is to calculate the surface roughness by measuring the spacing between the interference fringes of the two beams of light. However, the above method has high requirements for equipment, and due to the cumbersome steps, it is easily disturbed during the detection process, thereby affecting the detection accuracy.

With the development of artificial intelligence technology, there are studies combining light scattering with artificial intelligence models to measure roughness. For example, by establishing the relationship between the characteristic parameters of the speckle image and the surface roughness assessment parameters, efficient and non-destructive measurement of the surface roughness of the workpiece can be achieved. However, when extracting image features, this method often takes a lot of time to train the model due to the excessive number of features and the model's detection efficiency is not ideal.

Summary of the invention

In order to solve at least one of the above-mentioned technical problems, the present invention provides a method and system for detecting the surface roughness of an optical lens.

In a first aspect, the present invention provides a method for detecting the surface roughness of an optical lens, the method comprising:

irradiating the laser emitted by the laser device onto the surface of the optical lens to be tested to form a signal light, and using a spectroscopic device to separate the signal light into a first scattered light and a second scattered light with different wavelengths; and acquiring a first speckle image and a second speckle image after imaging based on the first scattered light and the second scattered light, respectively;

Performing image enhancement on the first speckle image and the second speckle image, wherein the image enhancement includes image enhancement and normalization processing;

Extracting texture features and grayscale mean features from the enhanced first speckle image, extracting texture features, grayscale standard deviation and grayscale root mean square features from the enhanced second speckle image, and inputting the extracted features of the first speckle image and the second speckle image into a support vector machine model for roughness detection, to obtain a first detection value and a second detection value;

The average value of the first detection value and the second detection value is calculated to obtain the surface roughness of the optical lens.

Preferably, the performing image enhancement on the first speckle image and the second speckle image comprises:

The first speckle image and the second speckle image are evenly divided into several sub-images respectively, and the flatness of each sub-image is calculated:

β _f =γcot ^-1 (δ _f )+ε,f=1,2,...,N

Where, β _f represents the flatness index of sub-image f; γ and ε represent the range factors controlling the flatness index; δ _f represents the standard deviation of sub-image f;

The optimal scale parameter of the filter is calculated based on the flatness index of the sub-image f:

Wherein, μ _f represents the optimal scale parameter of the filter corresponding to the sub-image f, μ _max and μ _min are the maximum and minimum values of the optimal scale parameter, max(β _f ) and min(β _f ) are the maximum and minimum values of the flatness index, respectively;

Calculate the weight of each sub-image:

b _f,i = |μ _f -μ _i |,i = 1, 2, 3

Wherein, μ ₁ , μ ₂ and μ ₃ are the minimum, middle and maximum values of the flatness index respectively; b _f,1 , b _f,2 , b _f,3 represent three different scales of sub-image f; w _f,c is the weight of sub-image f at scale c; i and c are scale indices;

Perform image enhancement and brightness normalization to obtain an enhanced image:

_Je (x,y)＝ _We (x,y)M(x,y)+ _le (x,y)(1-M(x,y))

Where, _Je (x,y) is the enhanced image of the speckle image, M(x,y) is the image of the speckle image after brightness normalization; _We (x,y) is the enhancement effect of the R, G and B color channels of the speckle image; _ΔCF is the compensation factor; l _e (x,y) is the brightness value of the e-th color channel of the optical lens surface image; e is the color channel index; is the convolution operator; _ge (x, y) is the effect of the e-th color channel of the optical lens surface image after being processed by the filter, h is the normalization factor; x and y are the horizontal and vertical coordinate indexes of the optical lens surface image respectively.

Preferably, when performing texture feature analysis on the enhanced first speckle image or the second speckle image, the method includes:

The texture features of the first speckle image or the second speckle image are extracted using the gray level co-occurrence matrix method:

Where E is energy, S is entropy, I is moment of inertia, L is correlation, H is inverse moment, and G _g is gray level. is the enhanced gray-level co-occurrence matrix; μ _x , μ _y are means; δ _x , δ _y are variances; x and y are the abscissa index and ordinate index of the optical lens surface image, respectively.

Preferably, before the extracted features of the first speckle image and the extracted features of the second speckle image are respectively input into a support vector machine model for roughness detection, the method further includes:

The Spearman correlation coefficient is used to conduct correlation analysis on the extracted texture features, and the inverse moment feature is eliminated according to the analysis results;

The Pearson correlation coefficient is used to perform correlation analysis on the extracted grayscale mean, grayscale standard deviation and grayscale root mean square, and the grayscale standard deviation feature is eliminated based on the analysis results.

Preferably, the expression of the support vector machine model is:

K( _xi ,x*)＝exp(-g‖xi- _x * ^‖2 )

Wherein, α ₁ and α ₂ represent Lagrange multipliers; _xi is the input sample; g is the kernel function parameter, K( _xi , x*) is the Gaussian radial basis kernel function; b is the bias value, and f(x*) represents the regression function value of the support vector machine.

Preferably, the method further comprises using mean absolute percentage error and root mean square error as evaluation indicators of the support vector machine model.

In a second aspect, the present invention further provides a system for detecting the surface roughness of an optical lens, the system comprising:

An image acquisition unit is used to irradiate the laser emitted by the laser device onto the surface of the optical lens to be tested to form a signal light, and use a spectroscopic device to separate the signal light into a first scattered light and a second scattered light with different wavelengths; based on the first scattered light and the second scattered light, respectively acquire a first speckle image and a second speckle image after imaging;

An image enhancement unit, configured to perform image enhancement on the first speckle image and the second speckle image, wherein the image enhancement includes image enhancement and normalization processing;

a feature extraction unit, configured to extract texture features and grayscale mean features from the enhanced first speckle image, extract texture features, grayscale standard deviation and grayscale root mean square features from the enhanced second speckle image, and input the extracted features of the first speckle image and the second speckle image into a support vector machine model for roughness detection, to obtain a first detection value and a second detection value;

The roughness detection unit is used to calculate the average value of the first detection value and the second detection value to obtain the surface roughness of the optical lens.

Preferably, the feature extraction unit is further used for:

The Spearman correlation coefficient is used to conduct correlation analysis on the extracted texture features, and the inverse moment feature is eliminated according to the analysis results;

The Pearson correlation coefficient is used to perform correlation analysis on the extracted grayscale mean, grayscale standard deviation and grayscale root mean square, and the grayscale standard deviation feature is eliminated based on the analysis results.

In a third aspect, the present invention further provides an electronic device comprising: a processor and a memory, wherein the memory is used to store computer program code, and the computer program code comprises computer instructions, and when the processor executes the computer instructions, the electronic device executes the method as described in the first aspect above and any possible implementation method thereof.

In a fourth aspect, the present invention further provides a computer-readable storage medium, wherein a computer program is stored in the computer-readable storage medium, wherein the computer program includes program instructions, and when the program instructions are executed by a processor of an electronic device, the processor executes the method as described in the first aspect above and any possible implementation thereof.

Compared with the prior art, the present invention has the following beneficial effects:

1) The present invention irradiates the laser emitted by the laser device on the surface of the optical lens to be tested to form signal light, and uses a spectroscopic device to separate the signal light into a first scattered light and a second scattered light with different wavelengths; based on the first scattered light and the second scattered light, a first speckle image and a second speckle image after imaging are respectively acquired; by using the roughness respectively detected by the first speckle image and the second speckle image and then removing the average value, the surface roughness of the optical lens can be detected based on different incident wavelengths, and compared with using a single light source, the accuracy of the detection result can be improved.

2) The present invention performs image enhancement on the first speckle image and the second speckle image, introduces a compensation factor to modify the logarithmic function of the MSR algorithm, and can better retain and enhance the detail information in the image. The optimal scale parameter is calculated based on the flatness index to determine the weight, which can effectively suppress the noise in the image, thereby better balancing the details and noise in the image. By normalizing the brightness of the image, the enhanced image can be made more balanced and natural, thereby improving the image quality.

3) The present invention extracts texture features and grayscale mean features from the enhanced first speckle image, extracts texture features, grayscale standard deviation and grayscale root mean square features from the enhanced second speckle image, and inputs the extracted features into the support vector machine model for roughness detection. The relationship between different image features and surface roughness can be studied at different wavelengths, thereby improving the accuracy of the detection results.

It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory only and are not restrictive of the present disclosure.

BRIEF DESCRIPTION OF THE DRAWINGS

In order to more clearly illustrate the technical solutions in the embodiments of the present invention or the background technology, the drawings required for use in the embodiments of the present invention or the background technology will be described below.

The drawings herein are incorporated into the specification and constitute a part of the specification. These drawings illustrate embodiments consistent with the present disclosure and are used to illustrate the technical solutions of the present disclosure together with the specification.

FIG1 is a schematic flow chart of a method for detecting surface roughness of an optical lens provided by an embodiment of the present invention;

FIG2 is a schematic structural diagram of a speckle image acquisition device provided by an embodiment of the present invention;

FIG3 is a schematic structural diagram of a system for detecting surface roughness of an optical lens provided by an embodiment of the present invention;

FIG. 4 is a schematic diagram of the hardware structure of an electronic device provided in an embodiment of the present invention.

Detailed ways

In order to enable those skilled in the art to better understand the scheme of the present invention, the technical scheme in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.

The terms "first", "second", etc. in the specification and claims of the present invention and the above-mentioned drawings are used to distinguish different objects, rather than to describe a specific order. In addition, the terms "including" and "having" and any variations thereof are intended to cover non-exclusive inclusions. For example, a process, method, system, product or device that includes a series of steps or units is not limited to the listed steps or units, but may optionally include steps or units that are not listed, or may optionally include other steps or units that are inherent to these processes, methods, products or devices.

The term "and/or" in this article is only a description of the association relationship of the associated objects, indicating that there can be three relationships. In addition, the term "at least one" in this article means any combination of at least two of any one or more of a plurality of, for example, including at least one of A, B, and C, can mean including any one or more elements selected from the set consisting of A, B, and C.

Reference to "embodiments" herein means that a particular feature, structure, or characteristic described in conjunction with the embodiments may be included in at least one embodiment of the present invention. The appearance of the phrase in various places in the specification does not necessarily refer to the same embodiment, nor is it an independent or alternative embodiment that is mutually exclusive of other embodiments. It is explicitly and implicitly understood by those skilled in the art that the embodiments described herein may be combined with other embodiments.

Please refer to Figure 1, which is a schematic flow chart of a method for detecting the surface roughness of an optical lens provided by an embodiment of the present invention. As shown in Figure 1, a method for detecting the surface roughness of an optical lens comprises the following steps:

S11, irradiating the laser emitted by the laser device onto the surface of the optical lens to be tested to form signal light, and using a spectroscopic device to separate the signal light into a first scattered light and a second scattered light with different wavelengths; based on the first scattered light and the second scattered light, respectively acquiring a first speckle image and a second speckle image after imaging.

The basic principle of measuring surface roughness by scattered light spots is that when a beam of light is incident on a rough surface, the reflected light is scattered in all directions in space due to the roughness of the surface, forming a scattering field. Therefore, by obtaining the scattered light spots and extracting the corresponding features, the surface roughness of the optical lens can be obtained by analyzing the features. In this embodiment, the first speckle image and the second speckle image obtained after scattering at different wavelengths are obtained.

Referring to FIG2 , FIG2 provides a measuring device for acquiring a speckle image. As shown in FIG2 , the measuring device comprises a laser 01 , a spectrometer 02 , a first camera 04 , a second camera 05 , and a stage 03 .

Specifically, the optical lens to be tested is mounted on a stage 03, and the stage 03 is movable. When the stage 03 moves, the angle of the incident laser emitted by the laser 01 can be changed. The laser 01 is used to emit a laser beam with a set tilt angle. After the laser beam passes through the spectrometer 02, it will be divided into beams of different angles and emitted to the surface of the optical lens to be tested; since the spectrometer 02 changes the phase of the incident light, the laser beam emitted by the laser 01 can be divided into incident lights of different wavelengths to obtain speckles under different imaging results. The first camera 04 and the second camera 05 are used to capture the first speckle image and the second speckle image, respectively. Preferably, the laser 01 adopts a collimated laser.

Therefore, in this embodiment, the wavelength of the incident light can be changed by the spectroscopic device, so as to obtain the corresponding speckle images under the conditions of incident light of different wavelengths. Compared with the measurement of incident light of a single wavelength, this embodiment can improve the accuracy of the detection result.

S12: Perform image enhancement on the first speckle image and the second speckle image, and normalize the brightness of the images.

It is easy to lose or distort image details when taking speckle images with a camera, and if the image is simply amplified, the image noise will be amplified, resulting in a decrease in the quality of the enhanced image. Therefore, in order to ensure the image quality and enable accurate extraction of image features later, this embodiment requires image enhancement and normalization.

In one embodiment, performing image enhancement on the first speckle image and the second speckle image comprises the following steps:

1) The first speckle image and the second speckle image are evenly divided into several sub-images respectively, and the flatness of each sub-image is calculated:

β _f =γcot ^-1 (δ _f )+ε,f=1,2,...,N

Where β _f represents the flatness index of sub-image f; γ and ε represent the range factors that control the flatness index; δ _f represents the standard deviation of sub-image f; β _f and δ _f satisfy an inverse relationship.

In order to ensure the effect of image enhancement, this embodiment first divides the image into regions, divides the speckle image into even regions, and then calculates the flatness of each region. Preferably, a flatness threshold can be set first, and then after calculating the flatness index of each region, the flatness index of the region is compared with the flatness threshold. If the difference exceeds, the region is considered to be a non-flat region, otherwise, it is considered to be a flat region.

2) Calculate the optimal scale parameter of the filter based on the flatness index of the sub-image f:

Wherein, μ _f represents the optimal scale parameter of the filter corresponding to the sub-image f, μ _max and μ _min are the maximum and minimum values of the optimal scale parameter, max(β _f ) and min(β _f ) are the maximum and minimum values of the flatness index, respectively;

3) Calculate the weight of each sub-image:

b _f,i = |μ _f -μ _i |,i = 1, 2, 3

Wherein, μ ₁ , μ ₂ and μ ₃ are the minimum, middle and maximum values of the flatness index respectively; b _f,1 , b _f,2 , b _f,3 represent three different scales of sub-image f; w _f,c is the weight of sub-image f at scale c; i and c are scale indices;

4) Perform image enhancement and brightness normalization to obtain an enhanced image:

_Je (x,y)＝ _We (x,y)M(x,y)+ _le (x,y)(1-M(x,y))

Where, _Je (x,y) is the enhanced image of the speckle image, M(x,y) is the image of the speckle image after brightness normalization; _We (x,y) is the enhancement effect of the R, G and B color channels of the speckle image; _ΔCF is the compensation factor; l _e (x,y) is the brightness value of the e-th color channel of the optical lens surface image; e is the color channel index; is the convolution operator; _ge (x, y) is the effect of the e-th color channel of the optical lens surface image after being processed by the filter, h is the normalization factor; x and y are the horizontal and vertical coordinate indexes of the optical lens surface image respectively.

Therefore, this embodiment introduces a compensation factor to correct the logarithmic function of the MSR algorithm, which can better retain and enhance the detail information in the image, calculate the optimal scale parameter based on the flatness index, and thus determine the weight, effectively suppress the noise in the image, better balance the details and noise in the image, and normalize the brightness of the image, so that the enhanced image is more balanced and natural, thereby improving the image quality.

S13, extracting texture features and grayscale mean features from the enhanced first speckle image, extracting texture features, grayscale standard deviation and grayscale root mean square features from the enhanced second speckle image, and inputting the extracted features into a support vector machine model for roughness detection, to obtain a first detection value and a second detection value.

It should be noted that when a rough surface is irradiated by a laser, the reflected light will form bright and dark spots on the observation plane. These randomly distributed bright and dark spots are called laser speckles. Laser speckles are usually described by the probability density function of light intensity. However, it is difficult to directly establish a surface roughness measurement model using the probability density function of light intensity. Therefore, it is necessary to extract features from the speckle image to study the direct relationship between image features and roughness.

Specifically, the texture features of the speckle image can be extracted by the grayscale co-occurrence matrix method and the grayscale difference statistics method, wherein the grayscale co-occurrence matrix method can extract five texture features, namely energy, entropy, moment of inertia, correlation and inverse moment. The grayscale difference statistics method can extract three features, namely, average value, contrast and entropy. It is also possible to extract the foreground pixel ratio of the binary image. Among them, the foreground pixel ratio of the binary image is defined as the ratio of the number of pixels with a pixel value of 1 to the total number of pixels in the binary image. Each extracted original feature parameter provides information about the speckle image, but not every original feature parameter has a good correlation with the surface roughness parameter. The more feature parameters there are, the more complicated the process of extracting the feature parameters is, and the higher the model data dimension is. Therefore, in this embodiment, the grayscale co-occurrence matrix method is mainly used to extract texture features and extract relevant features of the speckle grayscale image.

Specifically, the texture feature and grayscale mean feature extraction are performed on the enhanced first speckle image, including:

Where E is energy, S is entropy, I is moment of inertia, L is correlation, H is inverse moment, and G _g is gray level. is the enhanced gray-level co-occurrence matrix; μ _x , μ _y are means; δ _x , δ _y are variances; x and y are the abscissa index and ordinate index of the optical lens surface image, respectively.

In one embodiment, extracting the grayscale mean of the enhanced first speckle image includes:

Wherein, λ ₁ represents the grayscale mean of the first speckle image; N _x , N _y represent the number of pixels in the horizontal and vertical directions of the image, i, j represent the pixel points in the horizontal and vertical directions of the image, and I _g (i, j) represents the grayscale value of each pixel point in the grayscale image.

Furthermore, based on the enhanced gray level co-occurrence matrix, the texture features of the second speckle image are extracted in the same way as the texture features of the first speckle image, and five texture features, namely, energy E, entropy S, moment of inertia I, correlation L, and inverse moment H, are obtained respectively.

The grayscale standard deviation and grayscale root mean square are extracted from the enhanced second speckle image, including:

Wherein, λ ₂ represents the grayscale mean of the second speckle image; σ and ν represent the grayscale standard deviation and grayscale root mean square of the second speckle image.

Therefore, according to the features extracted in the above steps, the feature combinations of the first speckle image and the second speckle image can be constructed respectively. The feature combination of the first speckle image mainly includes five texture features of energy E, entropy S, moment of inertia I, correlation L, inverse moment H and one grayscale feature of grayscale mean λ ₁ ; while the feature combination of the second speckle image mainly includes five texture features of energy E, entropy S, moment of inertia I, correlation L, inverse moment H and two grayscale features of grayscale standard deviation σ and grayscale root mean square ν.

In order to further optimize the feature combination, this embodiment takes into account the correlation between features and removes features with weak correlation, thereby speeding up the training of the model and reducing the interference of redundant features on the model's ability to detect roughness.

In a preferred embodiment, before the extracted features of the first speckle image and the extracted features of the second speckle image are respectively input into a support vector machine model for roughness detection, the method further includes:

1) Use the Spearman correlation coefficient to perform correlation analysis on the extracted texture features, and eliminate the inverse moment features based on the analysis results.

The Spearman correlation coefficient is often used to measure the dependence between two variables. First, the two variables are rank-exchanged, and then the correlation between the two is calculated. If there are no repeated values in the data, when the two variables are completely monotonically correlated, the Spearman correlation coefficient is +1 or -1. Therefore, for the texture features of the first speckle image or the texture features of the second speckle image, the Spearman correlation coefficient can be used to calculate the correlation between the five features. Through calculation, it can be seen that the inverse difference moment feature has a weak correlation with other features, so this feature can be eliminated.

2) Use the Pearson correlation coefficient to perform correlation analysis on the extracted grayscale mean, grayscale standard deviation and grayscale root mean square, and eliminate the grayscale standard deviation feature based on the analysis results.

Furthermore, after removing the texture features, the grayscale features need to be subjected to correlation analysis. The grayscale feature in the first speckle image only has the grayscale mean, so no analysis is required. The grayscale features in the second speckle image include the grayscale standard deviation and the grayscale root mean square. Therefore, in this step, the correlation between the grayscale standard deviation, the grayscale root mean square and the grayscale mean is calculated by the Pearson correlation coefficient, so as to retain the features with stronger correlation. According to the calculation results, the grayscale root mean square is more strongly correlated with the grayscale mean. Therefore, in order to reduce the features, the grayscale standard deviation feature is removed here.

Through the above feature screening steps, it can be obtained that the features of the first speckle image include energy E, entropy S, moment of inertia I, correlation L, and grayscale mean λ ₁ ; the features of the second speckle image include energy E, entropy S, moment of inertia I, correlation L, and grayscale root mean square ν.

Finally, the features of the first speckle image and the features of the second speckle image are respectively input into the support vector machine model for roughness detection to obtain a first detection value and a second detection value.

It should be noted that the support vector machine (SVM) is a classification and regression algorithm suitable for small sample problems. By setting the linear regression function in high-dimensional space, the corresponding SVR model can be established. The specific expression is as follows:

K( _xi ,x*)＝exp(-g‖xi- _x *|| ² )

Wherein, α ₁ and α ₂ represent Lagrange multipliers; _xi is the input sample; g is the kernel function parameter, K( _xi , x*) is the Gaussian radial basis kernel function; b is the bias value, and f(x*) represents the regression function value of the support vector machine.

It can be understood that in the above steps, the process of feature extraction and feature screening is given. Finally, it is only necessary to input the finally screened features into the trained support vector machine model to obtain the roughness detection value. Therefore, when training the support vector machine model, it is also necessary to train based on the feature indicators used in the above embodiment. For example, several speckle images can be selected as training samples, and then for each sample image, the same features as in the above embodiment are extracted, and then the support vector machine is trained. After each training, the detection accuracy of the model needs to be evaluated. After the evaluation, the feature extraction step can be returned to screen and eliminate irrelevant features, and then the features used in the training model can be reselected, so that the detection accuracy of the model is better.

Preferably, the mean absolute percentage error and the root mean square error can be used as evaluation indicators of the support vector machine model. When the evaluation indicators do not meet the preset conditions, the model is trained by adjusting the input features until the indicators meet the preset conditions, and it is considered that the detection accuracy of the trained support vector machine model has met the requirements. Finally, the features of the first speckle image and the features of the second speckle image are respectively input into the trained support vector machine model for roughness detection, so that the first detection value and the second detection value can be obtained.

S14, calculating the average value of the first detection value and the second detection value to obtain the surface roughness of the optical lens.

In summary, this embodiment first uses a spectroscopic device to perform a spectroscopic operation on the light source emitted by the laser, and can obtain a speckle image generated by different wavelengths irradiating the surface of the optical lens to be tested, which can improve the accuracy of the detection result compared to the use of a single wavelength light source. By enhancing the image, the noise in the image can be effectively suppressed, thereby better balancing the details and noise in the image and improving the image quality; finally, for the speckle images obtained at different wavelengths, the Spearman correlation coefficient and the Pearson correlation coefficient are introduced to analyze the correlation of features, and finally different features are extracted by screening as inputs to the support vector machine model for roughness detection, and finally the roughness detection values obtained by the two are averaged to obtain the final surface roughness of the optical lens, which greatly improves the detection efficiency and the detection accuracy.

Referring to FIG. 3 , in one embodiment of the present invention, a system for detecting the surface roughness of an optical lens is further provided, the system comprising:

The image acquisition unit 100 is used to irradiate the laser emitted by the laser device onto the surface of the optical lens to be tested to form a signal light, and use a spectroscopic device to separate the signal light into a first scattered light and a second scattered light with different wavelengths; based on the first scattered light and the second scattered light, respectively acquire a first speckle image and a second speckle image after imaging;

An image enhancement unit 200, configured to perform image enhancement on the first speckle image and the second speckle image, wherein the image enhancement includes image enhancement and normalization processing;

The feature extraction unit 300 is used to extract texture features and grayscale mean features from the enhanced first speckle image, extract texture features, grayscale standard deviation and grayscale root mean square features from the enhanced second speckle image, and input the extracted features of the first speckle image and the second speckle image into a support vector machine model for roughness detection, to obtain a first detection value and a second detection value;

The roughness detection unit 400 is used to calculate the average value of the first detection value and the second detection value to obtain the surface roughness of the optical lens.

In one embodiment, the feature extraction unit 300 is further configured to:

The Spearman correlation coefficient is used to conduct correlation analysis on the extracted texture features, and the inverse moment feature is eliminated according to the analysis results;

The Pearson correlation coefficient is used to perform correlation analysis on the extracted grayscale mean, grayscale standard deviation and grayscale root mean square, and the grayscale standard deviation feature is eliminated based on the analysis results.

It can be understood that the functions or modules included in the system provided in this embodiment can be used to execute the method described in the above method embodiment. Its specific implementation can refer to the description of the above method embodiment. For the sake of brevity, it will not be repeated here.

The present invention also provides an electronic device, comprising: a processor, a sending device, an input device, an output device and a memory, wherein the memory is used to store computer program code, and the computer program code includes computer instructions. When the processor executes the computer instructions, the electronic device executes a method as described in any possible implementation manner.

The present invention also provides a computer-readable storage medium, in which a computer program is stored. The computer program includes program instructions. When the program instructions are executed by a processor of an electronic device, the processor executes a method as described in any possible implementation manner.

Please refer to FIG. 4 , which is a schematic diagram of the hardware structure of an electronic device provided by an embodiment of the present invention.

The electronic device 2 includes a processor 21, a memory 22, an input device 23, and an output device 24. The processor 21, the memory 22, the input device 23, and the output device 24 are coupled via a connector, and the connector includes various interfaces, transmission lines, or buses, etc., which are not limited in the embodiments of the present invention. It should be understood that in various embodiments of the present invention, coupling refers to mutual connection in a specific manner, including direct connection or indirect connection through other devices, for example, through various interfaces, transmission lines, buses, etc.

The processor 21 may be one or more graphics processing units (GPUs). When the processor 21 is a GPU, the GPU may be a single-core GPU or a multi-core GPU. Optionally, the processor 21 may be a processor group consisting of multiple GPUs, and the multiple processors are coupled to each other via one or more buses. Optionally, the processor may also be other types of processors, etc., which are not limited in the embodiments of the present invention.

The memory 22 can be used to store computer program instructions and various computer program codes including program codes for executing the scheme of the present invention. Optionally, the memory includes but is not limited to random access memory (RAM), read-only memory (ROM), erasable programmable read only memory (EPROM), or portable read only memory (CD-ROM), which is used for related instructions and data.

The input device 23 is used to input data and/or signals, and the output device 24 is used to output data and/or signals. The input device 23 and the output device 24 can be independent devices or an integrated device.

It is understandable that in the embodiment of the present invention, the memory 22 is not only used to store related instructions, and the embodiment of the present invention does not limit the specific data stored in the memory.

It is understandable that FIG4 only shows a simplified design of an electronic device. In practical applications, the electronic device may also include other necessary components, including but not limited to any number of input/output devices, processors, memories, etc., and all video analysis devices that can implement the embodiments of the present invention are within the protection scope of the present invention.

Those of ordinary skill in the art will appreciate that the units and algorithm steps of each example described in conjunction with the embodiments disclosed herein can be implemented in electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are performed in hardware or software depends on the specific application and design constraints of the technical solution. Professional and technical personnel can use different methods to implement the described functions for each specific application, but such implementation should not be considered to be beyond the scope of the present invention.

Those skilled in the art can clearly understand that, for the convenience and brevity of description, the specific working processes of the systems, devices and units described above can refer to the corresponding processes in the aforementioned method embodiments, and will not be repeated here. Those skilled in the art can also clearly understand that the descriptions of the various embodiments of the present invention have different focuses. For the convenience and brevity of description, the same or similar parts may not be repeated in different embodiments. Therefore, for parts not described or not described in detail in a certain embodiment, refer to the records of other embodiments.

In the several embodiments provided by the present invention, it should be understood that the disclosed systems, devices and methods can be implemented in other ways. For example, the device embodiments described above are only schematic. For example, the division of the units is only a logical function division. There may be other division methods in actual implementation, such as multiple units or components can be combined or integrated into another system, or some features can be ignored or not executed. Another point is that the mutual coupling or direct coupling or communication connection shown or discussed can be through some interfaces, indirect coupling or communication connection of devices or units, which can be electrical, mechanical or other forms.

The units described as separate components may or may not be physically separated, and the components shown as units may or may not be physical units, that is, they may be located in one place or distributed on multiple network units. Some or all of the units may be selected according to actual needs to achieve the purpose of the solution of this embodiment.

In addition, each functional unit in each embodiment of the present invention may be integrated into one processing unit, or each unit may exist physically separately, or two or more units may be integrated into one unit.

In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware or any combination thereof. When implemented by software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the process or function described in the embodiment of the present invention is generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted through the computer-readable storage medium. The computer instructions can be transmitted from a website site, computer, server or data center to another website site, computer, server or data center by wired (e.g., coaxial cable, optical fiber, digital subscriber line (digital subscriber line, DSL)) or wireless (e.g., infrared, wireless, microwave, etc.) mode. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that includes one or more available media integrated. The available medium may be a magnetic medium (eg, a floppy disk, a hard disk, a magnetic tape), an optical medium (eg, a digital versatile disc (DVD)), or a semiconductor medium (eg, a solid state disk (SSD)).

A person skilled in the art can understand that to implement all or part of the processes in the above-mentioned embodiments, the processes can be completed by a computer program to instruct the relevant hardware, and the program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above-mentioned method embodiments. The aforementioned storage medium includes: a read-only memory (ROM) or a random access memory (RAM), a magnetic disk or an optical disk, and other media that can store program codes.

Claims (10)

1. A method for detecting surface roughness of an optical lens, the method comprising:

Irradiating laser emitted by a laser device on the surface of an optical lens to be detected to form signal light, and separating the signal light into first scattered light and second scattered light with different wavelengths by using a light splitting device; acquiring a first speckle image and a second speckle image after imaging based on the first scattered light and the second scattered light, respectively;

performing image enhancement on the first speckle image and the second speckle image, wherein the image enhancement comprises image enhancement and normalization processing;

Extracting texture features and gray average features of the enhanced first speckle image, extracting texture features, gray standard deviation and gray root mean square features of the enhanced second speckle image, and respectively inputting the extracted features of the first speckle image and the extracted features of the second speckle image into a support vector machine model for roughness detection to obtain a first detection value and a second detection value;

and calculating the average value of the first detection value and the second detection value to obtain the surface roughness of the optical lens.

2. The method for detecting surface roughness of an optical lens of claim 1, wherein the image enhancement of the first speckle image and the second speckle image comprises:

Uniformly dividing the first speckle image and the second speckle image into a plurality of sub-images respectively, and calculating the flatness of each sub-image:

β_f＝γcot^-1(δ_f)+ε,f＝1,2,...,N

wherein β _f represents the flatness index of the sub-image f; gamma, epsilon represent the range factor controlling the flatness index; δ _f represents the standard deviation of the sub-image f;

Calculating an optimal scale parameter of the filter based on the flatness index of the sub-image f:

Wherein μ _f represents the optimal scale parameters of the corresponding filters of the sub-image f, μ _max and μ _min are the maximum and minimum values of the optimal scale parameters, respectively, and max (β _f) and min (β _f) are the maximum and minimum values of the flatness index, respectively;

Calculating the weight of each sub-image:

b_f,i＝|μ_f-μ_i|,i＝1,2,3

Wherein μ ₁、μ₂ and μ ₃ are the minimum, intermediate and maximum values of the flatness index, respectively; b _f,1、b_f,2,b_f,3 represents three different scales of the sub-image f; w _f,c is the weight of sub-image f on scale c; i and c are scale indices;

performing image enhancement and brightness normalization processing to obtain an enhanced image:

J_e(x,y)＝W_e(x,y)M(x,y)+l_e(x,y)(1-M(x,y))

Wherein J _e (x, y) is an enhanced image of the speckle image, and M (x, y) is an image of the speckle image subjected to brightness normalization; w _e (x, y) is the enhancement effect of the speckle image R, G and the B color channel; Δ _CF is the compensation factor; l _e (x, y) is the luminance value of the e-th color channel of the optical lens surface image; e is the color channel index; Is a convolution operator; g _e (x, y) is the effect of the e-th color channel of the optical lens surface image after being processed by the filter, and h is a normalization factor; x and y are the abscissa index and the ordinate index, respectively, of the optical lens surface image.

3. The method for detecting surface roughness of an optical lens of claim 1, wherein the texture feature of the enhanced first speckle image or the enhanced second speckle image comprises:

extracting texture features of the first speckle image or the second speckle image by using a gray level co-occurrence matrix method:

wherein E is energy; s is entropy; i is the moment of inertia; l is a correlation; h is the inverse moment, G _g is the number of gray levels; Is an enhanced gray level co-occurrence matrix; mu _x、μ_y is the mean; delta _x、δ_y is variance; x and y are the abscissa index and the ordinate index, respectively, of the optical lens surface image.

4. The method for detecting surface roughness of an optical lens as claimed in claim 1, wherein before the extracted features of the first speckle image and the extracted features of the second speckle image are respectively input to the support vector machine model for roughness detection, further comprising:

Carrying out correlation analysis on the extracted texture features by utilizing the spearman correlation coefficient, and eliminating inverse difference moment features according to analysis results;

and carrying out correlation analysis on the extracted gray average value, gray standard deviation and gray root mean square by using the pearson correlation coefficient, and eliminating gray standard deviation features according to analysis results.

5. The method for detecting surface roughness of an optical lens of claim 1, wherein the expression of the support vector machine model is:

K(x_i,x*)＝exp(-g||x_i-x*||²)

Wherein α ₁、α₂ represents a lagrange multiplier; x _i is the input sample; g is a kernel function parameter, and K (x _i, x) is a Gaussian radial basis kernel function; b is the bias value and f (x) represents the regression function value of the support vector machine.

6. The method for detecting surface roughness of an optical lens of claim 1, further comprising using the average absolute percentage error and the root mean square error as an evaluation index of a support vector machine model.

7. A system for detecting surface roughness of an optical lens, the system comprising:

An image acquisition unit for irradiating laser emitted by a laser device on the surface of an optical lens to be tested to form signal light, and separating the signal light into first scattered light and second scattered light with different wavelengths by using a light splitting device; acquiring a first speckle image and a second speckle image after imaging based on the first scattered light and the second scattered light, respectively;

An image enhancement unit for performing image enhancement on the first speckle image and the second speckle image, the image enhancement including image enhancement and normalization processing;

The feature extraction unit is used for extracting texture features and gray average features of the enhanced first speckle image, extracting texture features, gray standard deviation and gray root mean square features of the enhanced second speckle image, and respectively inputting the extracted features of the first speckle image and the extracted features of the second speckle image into the support vector machine model for roughness detection to obtain a first detection value and a second detection value;

And the roughness detection unit is used for calculating the average value of the first detection value and the second detection value to obtain the surface roughness of the optical lens.

8. The system for detecting surface roughness of an optical lens of claim 7, wherein the feature extraction unit is further configured to:

Carrying out correlation analysis on the extracted texture features by utilizing the spearman correlation coefficient, and eliminating inverse difference moment features according to analysis results;

and carrying out correlation analysis on the extracted gray average value, gray standard deviation and gray root mean square by using the pearson correlation coefficient, and eliminating gray standard deviation features according to analysis results.

9. An electronic device, comprising: a processor and a memory for storing computer program code comprising computer instructions which, when executed by the processor, the electronic device performs the method of detecting surface roughness of an optical lens as claimed in any one of claims 1 to 7.

10. A computer readable storage medium, characterized in that the computer readable storage medium has stored therein a computer program comprising program instructions which, when executed by a processor of an electronic device, cause the processor to perform the method of detecting the surface roughness of an optical lens according to any of claims 1 to 7.