CN116415474A

CN116415474A - Optical structure optimization method and device of lens group and electronic device

Info

Publication number: CN116415474A
Application number: CN202111632782.7A
Authority: CN
Inventors: 陈聪; 陈燚; 姜安琪; 李艳萍
Original assignee: Sunny Optical Zhejiang Research Institute Co Ltd
Current assignee: Sunny Optical Zhejiang Research Institute Co Ltd
Priority date: 2021-12-28
Filing date: 2021-12-28
Publication date: 2023-07-11

Abstract

The application relates to a method, a device, an electronic device and a storage medium for optimizing an optical structure of a lens group, wherein the method comprises the following steps: acquiring a plurality of optical structures of a lens group, wherein the MTF value and the main value parameter corresponding to each optical structure are different, and the total lens length is smaller than a preset first threshold value; PSF information under each optical structure is acquired, and a plurality of simulation images are generated according to the PSF information; inputting each simulation image into a trained image restoration model to obtain the threshold requirement of the trained image restoration model on the MTF value of the lens group and the uniformity requirement of the PSF value; and carrying out iterative optimization on the optical structure by adjusting the total lens length and/or the main value parameter so that the MTF value of the optimized optical structure meets the threshold requirement and the PSF value meets the uniformity requirement. According to the method and the device, the problem that the lens group cannot be thinned and high imaging quality can be achieved in the related art is solved, and the technical effect of improving the imaging quality of the lens group while thinning the lens group is achieved.

Description

Optical structure optimization method and device of lens group and electronic device

Technical Field

The present disclosure relates to the field of lens design technologies, and in particular, to a method and an apparatus for optimizing an optical structure of a lens group, an electronic device, and a storage medium.

Background

With the continuous development of smart phones, smart phones become thinner and more portable, but as pursuits of users for photographing image quality of the mobile phones are increasingly improved, the outsole photosensitive elements and the optical lens groups with more pieces are gradually applied to the smart phones, resulting in a gradual increase in thickness of the rear camera of the smart phones, thereby reducing portability of the smart phones.

At present, the optical structure of the lens group is directly optimized to achieve the object of thinning the lens group, for example, the structure, the surface shape or the material of the lens group are changed by utilizing optical design software, and the lens group is thinned by changing the refraction or diffraction characteristics of certain lenses.

At present, an effective solution is not proposed for solving the problem that the lens group in the related art cannot achieve both thinning and high imaging quality.

Disclosure of Invention

The embodiment of the application provides an optical structure optimization method, an optical structure optimization device, an electronic device and a storage medium of a lens group, which at least solve the problem that the lens group in the related art cannot achieve thinning and high imaging quality.

In a first aspect, an embodiment of the present application provides a method for optimizing an optical structure of a lens group, where the method includes: acquiring a plurality of preset optical structures of a lens group, wherein the MTF value and the main value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold; acquiring PSF information of the lens group under each optical structure, and generating a plurality of simulation images according to the PSF information, wherein the PSF information comprises PSFs under different object distances and different view angles, and each simulation image corresponds to one PSF under one optical structure; inputting each simulation image into a trained image restoration model to obtain the threshold requirement of the trained image restoration model on the MTF value of the lens group and the uniformity requirement of the PSF value; and iteratively optimizing the optical structure of the lens group by adjusting the total lens length and/or the main value parameter of the lens group so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement and the main value parameter meets the preset parameter index.

In some of these embodiments, obtaining the preset plurality of optical structures of the lens group includes: reducing the total lens length of the lens group to be smaller than the first threshold value; setting MTF values of different gradients, and obtaining main value parameters corresponding to each MTF value to obtain various optical structures of the lens group.

In some of these embodiments, iteratively optimizing the optical structure of the lens group by adjusting the lens full length and/or principal value parameters of the lens group comprises: taking an optical structure with the highest MTF value in the plurality of optical structures as a first optical structure; reducing the MTF value corresponding to the first optical structure, adjusting the main value parameter corresponding to the first optical structure based on the reduced MTF value, and performing iterative optimization on the first optical structure; determining that the first optical structure is optimized under the condition that main value parameters corresponding to the optimized first optical structure meet the parameter index, the MTF value meets the threshold requirement and the PSF value meets the uniformity requirement; under the condition that the main value parameter corresponding to the optimized first optical structure does not meet the parameter index and the MTF value is reduced to the lowest threshold value in the threshold value requirements, increasing the total lens length corresponding to the first optical structure to obtain a second optical structure; reducing the MTF value corresponding to the second optical structure, adjusting the main value parameter corresponding to the second optical structure based on the reduced MTF value, and performing iterative optimization on the second optical structure; and determining that the second optical structure is optimized under the condition that the main value parameter corresponding to the optimized second optical structure meets the parameter index, the MTF value meets the threshold requirement and the PSF value meets the uniformity requirement.

In some of these embodiments, obtaining PSF information for the lens group at each optical structure includes: acquiring a spectral response curve of an imaging sensor in the lens group; and acquiring PSF information of the lens group under each optical structure according to the optical structure and the spectral response curve, wherein the PSF information comprises PSFs of R channel, G channel and B channel under different object distances and different view angles.

In some of these embodiments, the method further comprises: taking an optical structure with the highest MTF value in the plurality of optical structures as a first optical structure; acquiring PSF information of the lens group under the first optical structure, wherein the PSF information comprises PSFs of an R channel, a G channel and a B channel under different object distances and different view angles; convolving each sample image in a preset training data set with each PSF respectively to obtain a plurality of degraded images corresponding to each sample image at different object distances; inputting each sample image and a plurality of degraded images corresponding to each sample image into a preset image restoration model, and optimizing parameter information of the image restoration model by using back propagation; and testing the image restoration performance of the image restoration model by using a preset test data set, and optimizing the parameter information of the image restoration model according to the image restoration performance of the image restoration model to obtain a trained image restoration model.

In some embodiments, inputting each simulated image into a trained image restoration model, and obtaining a threshold requirement of the trained image restoration model on the MTF value and a uniformity requirement of the PSF value of the lens group includes: inputting each simulation image into a trained image restoration model to obtain the image restoration performance of the trained image restoration model under each optical structure; determining threshold requirements of the trained image restoration model on the MTF value of the lens group according to whether the image restoration performance of the trained image restoration model under each optical structure meets preset target requirements; generating a first data set and a second data set according to the PSF information, wherein the first data set comprises simulation images corresponding to different view angles under a preset object distance, and the second data set comprises simulation images corresponding to different object distances under the preset view angle; and testing the image restoration performance of the trained image restoration model by using the first data set and the second data set corresponding to each optical structure to obtain the uniformity requirement of the trained image restoration model on the PSF value of the lens group.

In some of these embodiments, the threshold requirement comprises: the MTF value is larger than a preset second threshold value at a preset first frequency; the MTF value is larger than a preset third threshold value under a preset second frequency, wherein the first frequency is smaller than the second frequency; the uniformity requirement includes: under the first frequency and the preset object distance, the variance of the PSF convolution kernel sizes corresponding to the preset various view angles is smaller than a preset fourth threshold value; and under the first frequency and the preset view angle, the variance of the PSF convolution kernel sizes corresponding to the preset various object distances is smaller than a preset fifth threshold value.

In a second aspect, an embodiment of the present application provides an optical structure optimization apparatus for a lens group, the apparatus including: the first acquisition module is used for acquiring a plurality of preset optical structures of the lens group, wherein the MTF value and the main value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold value; the second acquisition module is used for acquiring PSF information of the lens group under each optical structure and generating a plurality of simulation images according to the PSF information, wherein each simulation image corresponds to the PSF information under one optical structure; the input module is used for inputting each simulation image into the trained image restoration model respectively to obtain the threshold requirement of the trained image restoration model on the MTF value and the uniformity requirement of the PSF value of the lens group; the optimization module is used for carrying out iterative optimization on the optical structure of the lens group by adjusting the lens total length and/or the main value parameter of the lens group so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement and the main value parameter meets the preset parameter index.

In a third aspect, embodiments of the present application further provide an electronic device, including a memory and a processor, where the memory stores a computer program, and the processor is configured to execute the computer program to perform the optical structure optimization method of the lens group according to the first aspect.

In a fourth aspect, an embodiment of the present application further provides a storage medium, where a computer program is stored, where the computer program, when executed by a processor, implements the optical structure optimization method of a lens group according to the first aspect.

Compared with the related art, the optical structure optimization method, the device, the electronic device and the storage medium of the lens group acquire a plurality of preset optical structures of the lens group, wherein the MTF value and the main value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold; acquiring PSF information of a lens group under each optical structure, and generating a plurality of simulation images according to the PSF information, wherein the PSF information comprises PSFs under different object distances and different view angles, and each simulation image corresponds to one PSF under one optical structure; inputting each simulation image into a trained image restoration model to obtain threshold requirements of the trained image restoration model on MTF values of the lens groups and uniformity requirements of PSF values; the optical structure of the lens group is subjected to iterative optimization by adjusting the lens total length and/or the main value parameters of the lens group, so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement, and the main value parameters meet the preset parameter indexes. The problem that the lens group cannot achieve thinning and high imaging quality in the related art is solved, and the technical effect of improving the imaging quality of the lens group while thinning the lens group is achieved.

The details of one or more embodiments of the application are set forth in the accompanying drawings and the description below to provide a more thorough understanding of the other features, objects, and advantages of the application.

Drawings

The accompanying drawings, which are included to provide a further understanding of the application and are incorporated in and constitute a part of this application, illustrate embodiments of the application and together with the description serve to explain the application and do not constitute an undue limitation to the application. In the drawings:

FIG. 1 is a flow chart of a method of optimizing an optical structure of a lens group according to an embodiment of the present application;

FIG. 2 is a schematic diagram of a network architecture of a convolutional neural network according to an embodiment of the present application;

fig. 3 is a graph showing contrast of imaging effects of a lens group according to an embodiment of the present application;

FIG. 4 is a block diagram of an optical configuration optimization device of a lens group according to an embodiment of the present application;

fig. 5 is a schematic structural diagram of an electronic device according to an embodiment of the present application.

Detailed Description

In order to make the objects, technical solutions and advantages of the present application more apparent, the present application is described and illustrated below with reference to the accompanying drawings and examples. It should be understood that the specific embodiments described herein are for purposes of illustration only and are not intended to limit the present application. All other embodiments, which can be made by one of ordinary skill in the art without undue burden on the person of ordinary skill in the art based on the embodiments provided herein, are intended to be within the scope of the present application. Moreover, it should be appreciated that while such a development effort might be complex and lengthy, it would nevertheless be a routine undertaking of design, fabrication, or manufacture for those of ordinary skill having the benefit of this disclosure, and thus should not be construed as having the benefit of this disclosure.

Reference in the specification to "an embodiment" means that a particular feature, structure, or characteristic described in connection with the embodiment may be included in at least one embodiment of the application. The appearances of such phrases in various places in the specification are not necessarily all referring to the same embodiment, nor are separate or alternative embodiments mutually exclusive of other embodiments. It is to be expressly and implicitly understood by those of ordinary skill in the art that the embodiments described herein can be combined with other embodiments without conflict.

Unless defined otherwise, technical or scientific terms used herein should be given the ordinary meaning as understood by one of ordinary skill in the art to which this application belongs. Reference to "a," "an," "the," and similar terms herein do not denote a limitation of quantity, but rather denote the singular or plural. The terms "comprising," "including," "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, article, or apparatus that comprises a list of steps or modules (elements) is not limited to only those steps or elements but may include other steps or elements not expressly listed or inherent to such process, method, article, or apparatus. The terms "connected," "coupled," and the like in this application are not limited to physical or mechanical connections, but may include electrical connections, whether direct or indirect. The term "plurality" as used herein means greater than or equal to two. "and/or" describes an association relationship of an association object, meaning that there may be three relationships, e.g., "a and/or B" may mean: a exists alone, A and B exist together, and B exists alone. The terms "first," "second," "third," and the like, as used herein, are merely distinguishing between similar objects and not representing a particular ordering of objects.

The present embodiment provides a method for optimizing an optical structure of a lens group, and fig. 1 is a flowchart of the method for optimizing an optical structure of a lens group according to an embodiment of the present application, as shown in fig. 1, where the method includes:

step S101, obtaining preset multiple optical structures of the lens group, wherein the MTF value and the principal value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold.

In the present embodiment, a modulation transfer function (Modulation Transfer Function, abbreviated as MTF) is a standard for analyzing the resolution of an optical system, and can be used to represent the characteristics of the optical system, and the larger the MTF value, the better the imaging quality of the optical system.

In the present embodiment, the principal value parameter is parameters such as an aperture size and a Field angle (FOV) Of the lens group.

In this embodiment, in the initial optical structure design stage, the lens group may be thinned first, that is, the total lens length (Total Track Length, abbreviated as TTL) of the lens group is reduced, the limitation on other main value parameters is relaxed, and a higher MTF optimization target is set, so as to obtain multiple optical structures, where in the multiple optical structures, besides the TTL meets the expectation, the aperture size and FOV cannot reach the standard before not being thinned, and the MTF value is slightly reduced compared with that before not being thinned, and further, the imaging quality of the lens group needs to be improved while the lens group is thinned through subsequent optimization.

Step S102, PSF information of the lens group under each optical structure is obtained, and a plurality of simulation images are generated according to the PSF information, wherein the PSF information comprises PSFs under different object distances and different view angles, and each simulation image corresponds to one PSF under one optical structure.

In this embodiment, the point spread function (Point Spread Function, abbreviated as PSF) is the light field distribution of the output image when the input object is a point light source.

In this embodiment, the PSF information includes PSFs of R channel, G channel and B channel under different object distances and different angles of view, and a preset high-definition image may be used to convolve with each PSF under one optical structure, so as to obtain multiple simulated images under different object distances under the optical structure or one simulated image under a certain object distance under the optical structure, which corresponds to the high-definition image, so as to obtain multiple simulated images under different object distances under each optical structure or one simulated image under a certain object distance under each optical structure, which corresponds to the high-definition image.

Step S103, inputting each simulation image into a trained image restoration model to obtain the threshold requirement of the trained image restoration model on the MTF value and the uniformity requirement of the PSF value of the lens group.

In this embodiment, the image restoration model may be a convolutional neural network model.

Convolutional neural networks (Convolutional Neural Networks, abbreviated as CNNs) are a type of feed-forward neural networks (Feedforward Neural Networks, abbreviated as FNNs) that contain convolutional computations and have a deep structure, and are one of the representative algorithms for deep learning. The convolutional neural network has connectivity and characteristic learning capability, so that corresponding features can be well learned from a large number of samples.

In this embodiment, the relationship between the sample image and the degraded image may be learned by the deep learning model, so as to achieve the image restoration effect of the image restoration model and improve the imaging quality of the lens group.

In one embodiment, the construction of the image restoration model may be implemented using a convolutional neural network of Res-Unet structure.

The Res-Unet is a U-net network with a residual structure, the U-net network is provided with an encoder and a decoder, a U-shaped structural symmetrical structure is formed, the U-net network is a widely applied network structure, a convolutional neural network with the Res-Unet structure can be built, reasonable loss function constraint is set, and a mapping model from degraded images with reduced definition to high-quality images is obtained through training, namely a trained image restoration model.

Step S104, iteratively optimizing the optical structure of the lens group by adjusting the lens total length and/or the main value parameters of the lens group, so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement, and the main value parameters meet the preset parameter indexes.

In the present embodiment, the image restoration model has two functions: one is to improve the sharpness of the image output by the imaging sensor in the lens group; secondly, the tolerance of MTF value reduction corresponding to the optical structure is given, and the degree of freedom of the optical structure design is increased; the image restoration model can put corresponding requirements on the low-high frequency MTF threshold value of the optical structure and the uniformity of the PSF value under different view angles and different object distances according to the image restoration performance of the optical structure under different MTF values and different PSFs, so as to guide the iterative optimization of the optical structure, and further obtain a restored image with higher quality in the final forward reasoning process.

The method can be used for gradually reducing the MTF value optimization target, optimizing the main value parameter and/or TTL of the lens group and performing iterative optimization on the optical structure of the lens group according to the threshold requirement of the image restoration model on the MTF value and the uniformity requirement on the PSF value, and further improving the imaging quality of the lens group while reducing the lens group, wherein the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement, and the main value parameter meets the preset parameter index.

In the related art, some technical solutions directly optimize the optical structure of the lens group to achieve the object of thinning the lens group, for example, by using optical design software, changing the structure, surface shape or material of the lens group, and thinning the lens group by changing the refraction or diffraction characteristics of some lenses, however, compared with the lens group not thinned, the lens group adopting these technical solutions has the problem of image quality loss, and is difficult to be suitable for equipment with high requirements on imaging quality.

In other technical schemes, aberration of an optical system of the lens group is removed by using a deep learning model, PSF with the aberration of the optical system is obtained, and then a training set is constructed according to the PSF, so that the aim of inhibiting the aberration of the optical system is fulfilled. The technical proposal avoids the complexity of increasing the hardware of the lens group, corrects the optical aberration by a software mode, and is beneficial to the miniaturization of the lens group. However, these solutions do not pay attention to feedback guidance on the optical structural design of the lens group, and cannot improve the imaging quality of the lens group while thinning the lens group.

In the above technical scheme, thinning and high imaging quality of the lens group cannot be considered, certain requirements on mounting precision and applicable equipment exist, and problems that application is difficult, various equipment cannot be adapted and the like exist.

According to the optical structure optimization method of the lens group, the optical structure design and the image restoration of the lens group are combined, the trained image restoration model is utilized, the feedback guidance is also carried out on the optical structure design while the definition and the quality of an image output by an imaging sensor of the lens group are improved, the optical structure and the image restoration model of the lens group are enabled to achieve end-to-end system optimization, the imaging quality of the lens group is improved while the lens group is thinned, and the optical structure optimization method can be applied to equipment with higher requirements on the image quality, such as mobile phones and tablet computers.

Through the steps S101 to S104, a plurality of preset optical structures of the lens group are obtained, wherein the MTF value and the main value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold; acquiring PSF information of a lens group under each optical structure, and generating a plurality of simulation images according to the PSF information, wherein the PSF information comprises PSFs under different object distances and different view angles, and each simulation image corresponds to one PSF under one optical structure; inputting each simulation image into a trained image restoration model to obtain threshold requirements of the trained image restoration model on MTF values of the lens groups and uniformity requirements of PSF values; the optical structure of the lens group is subjected to iterative optimization by adjusting the lens total length and/or the main value parameters of the lens group, so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement, and the main value parameters meet the preset parameter indexes. According to the method and the device, the problem that the lens group cannot be thinned and high imaging quality can be achieved in the related art is solved, and the technical effect of improving the imaging quality of the lens group while thinning the lens group is achieved.

In some of these embodiments, the threshold requirement comprises: the MTF value is larger than a preset second threshold value under a preset first frequency; the MTF value is larger than a preset third threshold value under a preset second frequency, wherein the first frequency is smaller than the second frequency; the uniformity requirements include: under the first frequency and the preset object distance, the variance of the PSF convolution kernel sizes corresponding to the preset various view angles is smaller than a preset fourth threshold value; and under the first frequency and the preset view angle, the variance of the PSF convolution kernel sizes corresponding to the preset various object distances is smaller than a preset fifth threshold value.

In this embodiment, the threshold requirements include a minimum threshold requirement for the MTF at low and high frequencies, taking the low frequency of 80lp/mm and the high frequency of 150lp/mm as an example, i.e., at a first frequency of 80lp/mm, the MTF value is required to be greater than the second threshold, and at a second frequency of 150lp/mm, the MTF value is required to be greater than the third threshold.

The uniformity requirement may include requiring that the variance of the PSF convolution kernel size at four field angles of 00, 03, 05, and 07 is less than a fourth threshold at the same object distance at a first frequency of 80 lp/mm; at a first frequency of 80lp/mm, the variance of the PSF convolution kernel size at four object distances of 30cm, 80cm, 150cm and infinity is required to be less than a fifth threshold at the same field angle.

In the above embodiment, the first threshold, the second threshold, the third threshold, the fourth threshold and the fifth threshold may be configured according to the actual needs of the lens group, and the first frequency, the second frequency, the multiple object distances and the multiple view angles may be selected by themselves.

The optical structure design of the lens group is fed back together through the threshold requirement of the trained image restoration model on the MTF value of the lens group and the uniformity requirement of the PSF value of the lens group, the aim that other main value parameters are not reduced while the lens group is thinned can be achieved through iterative optimization, and the imaging quality of the final lens group is slightly better than that of the lens group before thinning, so that the lens group is suitable for equipment with higher requirements on image quality, such as mobile phones, tablet computers and the like.

In some embodiments, the obtaining of the preset optical structures of the lens group may be achieved by:

step 1, the total lens length of the lens group is reduced to be smaller than a first threshold value.

And 2, setting MTF values of different gradients, and acquiring main value parameters corresponding to each MTF value to obtain various optical structures of the lens group.

In this embodiment, an optical structure with a reduced lens group height may be used as an initial design, that is, the TTL of the lens group is mainly limited, the limitation of other principal value parameters is relaxed, a higher MTF optimization target is set, the initial optimization is completed by using optical design software, and a first-version optical structure design is output, where the first-version optical structure design except the TTL meets a preset target, the principal value parameters such as the aperture size, FOV and the like cannot reach the standard before the lens group is not thinned, and the MTF is slightly reduced but not obvious compared with that before the lens group is not thinned.

In the above embodiment, the main value parameter may be continuously optimized for the first-version optical structure design, and MTF optimization targets of different gradients may be set, so as to obtain multiple optical structures with different gradient MTF values and different main value parameters.

In some embodiments, the iterative optimization of the optical structure of the lens group by adjusting the lens full length and/or principal value parameters of the lens group is achieved by:

and step 1, taking the optical structure with the highest MTF value in the plurality of optical structures as a first optical structure.

And step 2, reducing the MTF value corresponding to the first optical structure, and adjusting the main value parameter corresponding to the first optical structure based on the reduced MTF value to perform iterative optimization on the first optical structure.

And step 3, determining that the first optical structure is optimized under the condition that the main value parameter corresponding to the optimized first optical structure meets the parameter index, the MTF value meets the threshold value requirement and the PSF value meets the uniformity requirement.

And step 4, increasing the total lens length corresponding to the first optical structure to obtain a second optical structure under the condition that the main value parameter corresponding to the optimized first optical structure does not meet the parameter index and the MTF value is reduced to the lowest threshold value in the threshold value requirements.

And 5, reducing the MTF value corresponding to the second optical structure, and adjusting the main value parameter corresponding to the second optical structure based on the reduced MTF value to perform iterative optimization on the second optical structure.

And step 6, determining that the second optical structure is optimized under the condition that the main value parameter corresponding to the optimized second optical structure meets the parameter index, the MTF value meets the threshold value requirement and the PSF value meets the uniformity requirement.

In this embodiment, the optical structure with the highest MTF value in the multiple optical structures may be used as the first optical structure for iterative optimization, and any one of the multiple optical structures may be selected as the first optical structure for iterative optimization.

In this embodiment, according to the threshold requirement of the trained image restoration model on the MTF value of the lens group and the uniformity requirement of the PSF value of the lens group, the limitation of parameters such as curvature of field, aberration and chromatic aberration of the first optical structure can be relaxed, the MTF value corresponding to the first optical structure can be reduced, and the main value parameters such as aperture size and FOV of the first optical structure can be optimized until the main value parameters meet the parameter index, or the MTF value has been reduced to the lowest threshold value of the threshold requirements.

In the above embodiment, if the main value parameter corresponding to the first optical structure has reached the standard, the MTF value meets the threshold requirement, and the PSF value meets the uniformity requirement, the first optical structure is confirmed to directly complete the iterative optimization process; if the main value parameter corresponding to the first optical structure does not reach the standard, the limitation on the TTL of the first optical structure can be gradually relaxed to obtain a second optical structure, the iterative optimization is performed on the second optical structure until the main value parameter of the second optical structure reaches the standard, the MTF value just meets the threshold value requirement, the PSF value just meets the uniformity requirement, and the TTL corresponding to the second optical structure at the moment, namely the limit height which can be compressed by the current lens group, is completed.

In the above embodiment, after the iterative optimization of the first optical structure or the second optical structure is completed, the optimized first optical structure or second optical structure may be fine-tuned according to the conventional structural design manner under the condition that the main value parameter is unchanged, so as to improve the applicability of the lens group, and improve the imaging quality of the lens group while thinning the lens group.

In some of these embodiments, obtaining PSF information for a lens group under each optical structure is achieved by:

Step 1, acquiring a spectral response curve of an imaging sensor in a lens group.

And 2, acquiring PSF information of the lens group under each optical structure according to the optical structure and the spectral response curve, wherein the PSF information comprises PSFs of an R channel, a G channel and a B channel under different object distances and different angles of view.

In this embodiment, according to parameters such as an optical structure of the lens group and a spectral response curve of the imaging sensor, a light path propagation model such as a diffraction model and a geometric optical imaging model can be utilized, so as to obtain a process that the imaging sensor calculates PSF information of the optical system in a forward direction, and further calculate PSFs of an R channel, a G channel and a B channel in all view angles corresponding to each optical structure of the lens group under different object distances.

In some of these embodiments, the method further comprises the steps of:

And 2, acquiring PSF information of the lens group under the first optical structure, wherein the PSF information comprises PSFs of an R channel, a G channel and a B channel under different object distances and different view angles.

And step 3, respectively convolving each sample image in the preset training data set with each PSF to obtain a plurality of degraded images corresponding to each sample image under different object distances.

And 4, inputting each sample image and a plurality of degraded images corresponding to each sample image into a preset image restoration model, and optimizing parameter information of the image restoration model by using back propagation.

And 5, testing the image restoration performance of the image restoration model by using a preset test data set, and optimizing the parameter information of the image restoration model according to the image restoration performance of the image restoration model to obtain a trained image restoration model.

In this embodiment, for the first optical structure, PSFs of different angles of view corresponding to object distances may be obtained at near, intermediate and far distances, and the PSFs of different angles of view may be convolved with the sample image, so as to obtain degraded images at the near, intermediate and far distances, and the degraded images at different object distances and the sample image may be input into an image restoration model for Fine tuning (Fine Tune) training, so as to improve the performance of the image restoration model, further improve the image restoration quality of the image restoration model, and improve the imaging quality of the lens group.

In this embodiment, the degradation imaging process can be represented by convolution operation: y=x×a, where x represents convolution operation, y represents a degraded image, x is a sample image, and a represents a PSF.

In this embodiment, a plurality of sample images may be used, for example, a high-definition RGB data set with a resolution of 4K greater than 2000 is used to respectively convolve with each PSF, and a plurality of degraded images are generated through simulation.

In this embodiment, the sample image may be convolved with each PSF, and different angles of view may be split by using annular segmentation and/or dicing, and a preset white gaussian noise may be added to the generated blurred image, to obtain degraded images corresponding to the sample image at different object distances.

In this embodiment, in order to simulate the incompletely corrected blur in the optical system, the sample image and the PSF may be convolved, and then a preset white gaussian noise is added to the generated blurred image to simulate the optical path noise, where the standard deviation of the white gaussian noise may follow a value between 0 and 3, and further, after obtaining the degraded image corresponding to the sample image at different object distances, the degraded image may be normalized, and normalized, so as to facilitate the subsequent training model process.

In this embodiment, each sample image and a plurality of degraded images corresponding to each sample image at different object distances may be used as training data pairs to be input into a preset image restoration model, or the training data pairs may be diced to obtain a plurality of image blocks, and each image block is input into the image restoration model after data amplification.

In this embodiment, the data of each image block may be amplified by means of horizontal/vertical overturn, rotation, scaling, clipping, shearing, translation, contrast adjustment, color dithering, noise addition, and the like, so as to increase the training dataset of the image restoration model, so that the training dataset is diversified as much as possible, and the generalization capability of the image restoration model is improved.

In this embodiment, the test data set includes a simulated image under an untrained object distance, the image restoration performance of the image restoration model is tested by the test data set, and the parameter information of the image restoration model is adjusted according to the image restoration performance of the image restoration model, including the network structure, the structure of the training data set, and the constraint of the loss function, until the restored image output by the image restoration model meets the preset requirement, so as to obtain the trained image restoration model.

In this embodiment, the training data set is input to the image restoration model, global minimization is performed on the loss function, the parameter information of the image restoration model is optimized according to the back propagation function of the image restoration model, the constructed image restoration model can directly output a corresponding restoration image after the degraded image is input, meanwhile, the threshold value requirement can be provided for the MTF value of the lens group based on the image restoration performance of the image restoration model, the uniformity requirement is provided for the PSF value of the lens group to guide iterative optimization of the optical structural design, thus obtaining a restoration image with higher quality in the final forward reasoning process, and feedback guidance is performed on the optical structural design while improving the definition and quality of the image output by the imaging sensor of the lens group, so that the end-to-end system optimization is achieved between the optical structure of the lens group and the image restoration model, and the imaging quality of the lens group is improved while the lens group is thinned, and the imaging quality of the lens group is applicable to devices with higher requirements for the image quality such as mobile phones and tablet computers.

Fig. 2 is a schematic diagram of a network structure of a convolutional neural network according to an embodiment of the present application, and as shown in fig. 2, an image restoration model is constructed by using a convolutional neural network with a Res-Unet structure, where the convolutional neural network includes a plurality of convolutional layers, deconvolution layers, and residual blocks.

The convolution layer is used to extract different features of the image input to the image restoration model, wherein the low convolution layer may only extract some low-level features such as levels of edges, lines, and angles, and more layers of the network can iteratively extract more complex features from the low-level features.

The residual structure provides more debugging performance for the image restoration model, the image restoration model can control the superposition proportion of the last layer by adjusting the weight factor K, so that the definition of the output image is controlled, and the specific formula is as follows:

Output＝Input+K*Layer _out 。

in this embodiment, an actual lens group may be used to capture point light sources under different angles of view, so as to obtain PSFs of the lens group under different angles of view, and obtain PSFs under different angles of view corresponding to object distances under short distance, intermediate distance and long distance, and the PSFs under different object distances and different angles of view are convolved by using the high-definition Raw image in a blocking manner, so as to obtain corresponding first training data pairs, where the first training data pairs include the high-definition Raw image and a degraded image formed after convolution.

The high-definition displays under the short distance, the medium distance and the long distance can be shot by using an actual lens group, a corresponding real-shot Raw image and a projection RGB image are obtained, the Raw image degradation operation is carried out on the projection RGB image, the position, the brightness and the color of the projection RGB image are matched with the real-shot Raw image to generate a GT (true sample) image, and the GT image and the real-shot Raw image are used as a second training data pair.

Inputting the first training data pair and the second training data pair into the image restoration model, and performing Fine tuning (Fine Tune) training on the image restoration model to improve the performance of the image restoration model, so as to further improve the image restoration quality of the image restoration model and improve the imaging quality of the under-screen camera.

In some embodiments, each simulation image is input into a trained image restoration model, and the threshold requirement of the trained image restoration model on the MTF value and the uniformity requirement of the PSF value of the lens group are achieved through the following steps:

and step 1, respectively inputting each simulation image into a trained image restoration model to obtain the image restoration performance of the trained image restoration model under each optical structure.

And 2, determining the threshold requirement of the trained image restoration model on the MTF value of the lens group according to whether the image restoration performance of the trained image restoration model under each optical structure meets the preset target requirement.

And 3, generating a first data set and a second data set according to PSF information, wherein the first data set comprises simulation images corresponding to different view angles under a preset object distance, and the second data set comprises simulation images corresponding to different object distances under the preset view angle.

And 4, testing the image restoration performance of the trained image restoration model by using the first data set and the second data set corresponding to each optical structure to obtain the uniformity requirement of the trained image restoration model on the PSF value of the lens group.

In this embodiment, the trained image restoration model may be trained according to PSF information corresponding to optical structures of different gradient MTF values, and the threshold requirement of the trained image restoration model on the MTF value of the lens group may be determined according to whether the restoration image output by the trained image restoration model meets a preset target requirement, where the target requirement may be that the restoration image of the trained image restoration model is better than the image output by the lens before thinning in subjective evaluation and objective evaluation.

In this embodiment, the trained image restoration model may be trained and validated by using PSF information corresponding to each optical structure at different object distances and different angles of view to generate simulated images with different PSF uniformity degrees at different angles of view at a single object distance; the simulation images with different PSF uniformity degrees of different object distances under a single field angle can be generated to train and verify the effect of the trained image restoration model, so that the PSF uniformity requirement of the trained image restoration model on the lens group is obtained.

Through the embodiment, the trained image restoration model is subjected to the threshold value requirement on the MTF value of the lens group and the uniformity requirement on the PSF value of the lens group, and simultaneously the iterative optimization design of the optical structure is fed back, so that a restoration image with higher quality is obtained in the final forward reasoning process, the definition and the quality of an image output by an imaging sensor of the lens group are improved, and simultaneously the feedback guidance is carried out on the optical structure design, so that the end-to-end system optimization of the optical structure of the lens group and the image restoration model is achieved, and the imaging quality of the lens group is improved while the lens group is thinned.

Fig. 3 is a comparison chart of imaging effects of a lens group according to an embodiment of the present application, in the embodiment, the optical structure optimization method of the lens group of the present application can design a thinner optical lens group under the requirement that high specification parameters such as an outsole photosensitive element, a large aperture, a small TTL and the like are commonly adopted in a current smart phone, and simultaneously obtain higher imaging quality, in a specific embodiment, the optical structure optimization method of the lens group of the present application can reduce the TTL of the lens group by about 20% while keeping other principal value parameters of the lens group of the smart phone unchanged, as shown in fig. 3, the image on the left side is the imaging effect of the lens group which is finished by adopting the optical structure optimization method of the lens group of the present application, and the image on the right side is the imaging effect of the lens group which is not finished by adopting the optical structure optimization method of the lens group of the present application, so that the imaging effect of the lens group which is finished by adopting the optical structure optimization method of the lens group of the present application is slightly superior to the imaging key effect of the lens group which is not finished by adopting the lens group which is not finished.

The present embodiment provides an optical structure optimization device of a lens group, and fig. 4 is a block diagram of the optical structure optimization device of the lens group according to an embodiment of the present application, as shown in fig. 4, where the device includes: a first obtaining module 41, configured to obtain preset multiple optical structures of the lens group, where an MTF value and a principal value parameter of the lens group corresponding to each optical structure are different, and a total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold; a second obtaining module 42, configured to obtain PSF information of the lens group under each optical structure, and generate a plurality of simulated images according to the PSF information, where the PSF information includes PSFs under different object distances and different angles of view, and each simulated image corresponds to one PSF under one optical structure; the input module 43 is configured to input each simulation image into a trained image restoration model, so as to obtain a threshold requirement of the trained image restoration model on an MTF value and a uniformity requirement of a PSF value of the lens group; the optimization module 44 is configured to iteratively optimize the optical structure of the lens group by adjusting the lens total length and/or the principal value parameter of the lens group, so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement, and the principal value parameter meets the preset parameter index.

In some of these embodiments, the first obtaining module 41 is further configured to reduce the total lens length of the lens group to be less than the first threshold; setting MTF values of different gradients, and obtaining main value parameters corresponding to each MTF value to obtain various optical structures of the lens group.

In some of these embodiments, the optimization module 44 is further configured to take as the first optical structure the optical structure having the highest MTF value of the plurality of optical structures; reducing the MTF value corresponding to the first optical structure, adjusting the main value parameter corresponding to the first optical structure based on the reduced MTF value, and performing iterative optimization on the first optical structure; determining that the first optical structure is optimized under the condition that the main value parameter corresponding to the optimized first optical structure meets the parameter index, the MTF value meets the threshold value requirement and the PSF value meets the uniformity requirement; under the condition that the main value parameter corresponding to the optimized first optical structure does not meet the parameter index and the MTF value is reduced to the lowest threshold value in the threshold value requirements, increasing the total lens length corresponding to the first optical structure to obtain a second optical structure; reducing the MTF value corresponding to the second optical structure, adjusting the main value parameter corresponding to the second optical structure based on the reduced MTF value, and performing iterative optimization on the second optical structure; and under the condition that the main value parameter corresponding to the optimized second optical structure meets the parameter index, the MTF value meets the threshold value requirement and the PSF value meets the uniformity requirement, determining that the second optical structure is optimized.

In some of these embodiments, the second acquisition module 42 is further configured to acquire a spectral response curve of the imaging sensors in the lens group; PSF information of the lens group under each optical structure is obtained according to the optical structure and the spectral response curve, wherein the PSF information comprises PSFs of R channel, G channel and B channel under different object distances and different view angles.

In some embodiments, the apparatus further comprises a training module configured to take an optical structure with a highest MTF value of the plurality of optical structures as the first optical structure; acquiring PSF information of a lens group under a first optical structure, wherein the PSF information comprises PSFs of an R channel, a G channel and a B channel under different object distances and different angles of view; convolving each sample image in a preset training data set with each PSF respectively to obtain a plurality of degraded images corresponding to each sample image under different object distances; inputting each sample image and a plurality of degraded images corresponding to each sample image into a preset image restoration model, and optimizing parameter information of the image restoration model by using back propagation; and testing the image restoration performance of the image restoration model by using a preset test data set, and optimizing the parameter information of the image restoration model according to the image restoration performance of the image restoration model to obtain a trained image restoration model.

In some of these embodiments, the input module 43 is further configured to input each simulated image into the trained image restoration model separately, resulting in image restoration performance of the trained image restoration model for each optical structure; determining threshold requirements of the trained image restoration model on MTF values of the lens group according to whether the image restoration performance of the trained image restoration model under each optical structure meets preset target requirements; generating a first data set and a second data set according to PSF information, wherein the first data set comprises simulation images corresponding to different view angles under a preset object distance, and the second data set comprises simulation images corresponding to different object distances under the preset view angle; and testing the image restoration performance of the trained image restoration model by using the first data set and the second data set corresponding to each optical structure to obtain the uniformity requirement of the trained image restoration model on the PSF value of the lens group.

It should be noted that, specific examples in this embodiment may refer to examples described in the foregoing embodiments and alternative implementations, and this embodiment is not repeated herein.

The present embodiment also provides an electronic device, and fig. 5 is a schematic diagram of a hardware structure of the electronic device according to an embodiment of the present application, and as shown in fig. 5, the electronic device includes a memory 504 and a processor 502, where the memory 504 stores a computer program, and the processor 502 is configured to execute the computer program to perform steps in any one of the method embodiments described above.

In particular, the processor 502 may include a Central Processing Unit (CPU), or an application specific integrated circuit (Application Specific Integrated Circuit, abbreviated as ASIC), or may be configured to implement one or more integrated circuits of embodiments of the present application.

Memory 504 may include, among other things, mass storage for data or instructions. By way of example, and not limitation, memory 504 may comprise a Hard Disk Drive (HDD), floppy Disk Drive, solid state Drive (Solid State Drive, SSD), flash memory, optical Disk, magneto-optical Disk, tape, or Universal Serial Bus (USB) Drive, or a combination of two or more of the foregoing. The memory 504 may include removable or non-removable (or fixed) media, where appropriate. The memory 504 may be internal or external to the optical configuration optimization device of the lens group, where appropriate. In a particular embodiment, the memory 504 is a Non-Volatile (Non-Volatile) memory. In a particular embodiment, the Memory 504 includes Read-Only Memory (ROM) and random access Memory (Random Access Memory, RAM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (Programmable Read-Only Memory, abbreviated PROM), an erasable PROM (Erasable Programmable Read-Only Memory, abbreviated EPROM), an electrically erasable PROM (Electrically Erasable Programmable Read-Only Memory, abbreviated EEPROM), an electrically rewritable ROM (Electrically Alterable Read-Only Memory, abbreviated EAROM), or a FLASH Memory (FLASH), or a combination of two or more of these. The RAM may be Static Random-Access Memory (SRAM) or dynamic Random-Access Memory (Dynamic Random Access Memory DRAM), where the DRAM may be a fast page mode dynamic Random-Access Memory (Fast Page Mode Dynamic Random Access Memory FPMDRAM), extended data output dynamic Random-Access Memory (Extended Date Out Dynamic Random Access Memory EDODRAM), synchronous dynamic Random-Access Memory (Synchronous Dynamic Random-Access Memory SDRAM), or the like, as appropriate.

Memory 504 may be used to store or cache various data files that need to be processed and/or used for communication, as well as possible computer program instructions for execution by processor 502.

The processor 502 implements the optical structure optimization method of any one of the lens groups of the above embodiments by reading and executing computer program instructions stored in the memory 504.

Optionally, the electronic apparatus may further include a transmission device 506 and an input/output device 508, where the transmission device 506 is connected to the processor 502 and the input/output device 508 is connected to the processor 502.

Alternatively, in this embodiment, the processor 502 may be configured to execute the following steps by a computer program:

s1, acquiring preset multiple optical structures of a lens group, wherein the MTF value and the principal value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold value.

S2, PSF information of the lens group under each optical structure is obtained, and a plurality of simulation images are generated according to the PSF information, wherein the PSF information comprises PSFs under different object distances and different view angles, and each simulation image corresponds to one PSF under one optical structure.

S3, inputting each simulation image into a trained image restoration model to obtain the threshold requirement of the trained image restoration model on the MTF value and the uniformity requirement of the PSF value of the lens group.

S4, performing iterative optimization on the optical structure of the lens group by adjusting the total lens length and/or the main value parameters of the lens group, so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement, and the main value parameters meet the preset parameter indexes.

In addition, in combination with the optical structure optimization method of the lens group in the above embodiment, the embodiment of the application may provide a storage medium to be implemented. The storage medium has a computer program stored thereon; the computer program, when executed by a processor, implements the optical structure optimization method of any one of the lens groups of the above embodiments.

It should be understood by those skilled in the art that the technical features of the above embodiments may be combined in any manner, and for brevity, all of the possible combinations of the technical features of the above embodiments are not described, however, they should be considered as being within the scope of the description provided herein, as long as there is no contradiction between the combinations of the technical features.

The foregoing examples merely represent several embodiments of the present application, the description of which is more specific and detailed and which should not be construed as limiting the scope of the present application in any way. It should be noted that it would be apparent to those skilled in the art that various modifications and improvements could be made without departing from the spirit of the present application, which would be within the scope of the present application. Accordingly, the scope of protection of the present application shall be subject to the appended claims.

Claims

1. A method for optimizing an optical structure of a lens group, the method comprising:

acquiring a plurality of preset optical structures of a lens group, wherein the MTF value and the main value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold;

acquiring PSF information of the lens group under each optical structure, and generating a plurality of simulation images according to the PSF information, wherein the PSF information comprises PSFs under different object distances and different view angles, and each simulation image corresponds to one PSF under one optical structure;

inputting each simulation image into a trained image restoration model to obtain the threshold requirement of the trained image restoration model on the MTF value of the lens group and the uniformity requirement of the PSF value;

And iteratively optimizing the optical structure of the lens group by adjusting the total lens length and/or the main value parameter of the lens group so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement and the main value parameter meets the preset parameter index.

2. The method of optimizing an optical structure of a lens group according to claim 1, wherein obtaining a preset plurality of optical structures of the lens group comprises:

reducing the total lens length of the lens group to be smaller than the first threshold value;

setting MTF values of different gradients, and obtaining main value parameters corresponding to each MTF value to obtain various optical structures of the lens group.

3. The method of optimizing an optical structure of a lens group according to claim 1, wherein iteratively optimizing the optical structure of the lens group by adjusting a total lens length and/or a principal value parameter of the lens group comprises:

taking an optical structure with the highest MTF value in the plurality of optical structures as a first optical structure;

reducing the MTF value corresponding to the first optical structure, adjusting the main value parameter corresponding to the first optical structure based on the reduced MTF value, and performing iterative optimization on the first optical structure;

Determining that the first optical structure is optimized under the condition that main value parameters corresponding to the optimized first optical structure meet the parameter index, the MTF value meets the threshold requirement and the PSF value meets the uniformity requirement;

under the condition that the main value parameter corresponding to the optimized first optical structure does not meet the parameter index and the MTF value is reduced to the lowest threshold value in the threshold value requirements, increasing the total lens length corresponding to the first optical structure to obtain a second optical structure;

reducing the MTF value corresponding to the second optical structure, adjusting the main value parameter corresponding to the second optical structure based on the reduced MTF value, and performing iterative optimization on the second optical structure;

and determining that the second optical structure is optimized under the condition that the main value parameter corresponding to the optimized second optical structure meets the parameter index, the MTF value meets the threshold requirement and the PSF value meets the uniformity requirement.

4. The method of optimizing an optical structure of a lens group according to claim 1, wherein acquiring PSF information of the lens group at each optical structure comprises:

acquiring a spectral response curve of an imaging sensor in the lens group;

And acquiring PSF information of the lens group under each optical structure according to the optical structure and the spectral response curve, wherein the PSF information comprises PSFs of R channel, G channel and B channel under different object distances and different view angles.

5. The method of optimizing an optical structure of a lens group according to claim 1, further comprising:

acquiring PSF information of the lens group under the first optical structure, wherein the PSF information comprises PSFs of an R channel, a G channel and a B channel under different object distances and different view angles;

convolving each sample image in a preset training data set with each PSF respectively to obtain a plurality of degraded images corresponding to each sample image under different object distances;

inputting each sample image and a plurality of degraded images corresponding to each sample image into a preset image restoration model, and optimizing parameter information of the image restoration model by using back propagation;

and testing the image restoration performance of the image restoration model by using a preset test data set, and optimizing the parameter information of the image restoration model according to the image restoration performance of the image restoration model to obtain a trained image restoration model.

6. The method of optimizing an optical structure of a lens group according to claim 1, wherein inputting each simulation image into a trained image restoration model to obtain a threshold requirement of the trained image restoration model on an MTF value and a uniformity requirement of a PSF value of the lens group comprises:

inputting each simulation image into a trained image restoration model to obtain the image restoration performance of the trained image restoration model under each optical structure;

determining threshold requirements of the trained image restoration model on the MTF value of the lens group according to whether the image restoration performance of the trained image restoration model under each optical structure meets preset target requirements;

generating a first data set and a second data set according to the PSF information, wherein the first data set comprises simulation images corresponding to different view angles under a preset object distance, and the second data set comprises simulation images corresponding to different object distances under the preset view angle;

and testing the image restoration performance of the trained image restoration model by using the first data set and the second data set corresponding to each optical structure to obtain the uniformity requirement of the trained image restoration model on the PSF value of the lens group.

7. The optical structure optimization method of a lens group according to any one of claims 1 to 6, wherein the threshold requirement includes: the MTF value is larger than a preset second threshold value at a preset first frequency; the MTF value is larger than a preset third threshold value under a preset second frequency, wherein the first frequency is smaller than the second frequency;

the uniformity requirement includes: under the first frequency and the preset object distance, the variance of the PSF convolution kernel sizes corresponding to the preset various view angles is smaller than a preset fourth threshold value; and under the first frequency and the preset view angle, the variance of the PSF convolution kernel sizes corresponding to the preset various object distances is smaller than a preset fifth threshold value.

8. An optical structure optimization device for a lens group, the device comprising:

the first acquisition module is used for acquiring a plurality of preset optical structures of the lens group, wherein the MTF value and the main value parameter of the lens group corresponding to each optical structure are different, and the total lens length of the lens group corresponding to each optical structure is smaller than a preset first threshold value;

the second acquisition module is used for acquiring PSF information of the lens group under each optical structure and generating a plurality of simulation images according to the PSF information, wherein each simulation image corresponds to the PSF information under one optical structure;

The input module is used for inputting each simulation image into the trained image restoration model respectively to obtain the threshold requirement of the trained image restoration model on the MTF value and the uniformity requirement of the PSF value of the lens group;

the optimization module is used for carrying out iterative optimization on the optical structure of the lens group by adjusting the lens total length and/or the main value parameter of the lens group so that the MTF value corresponding to the optimized optical structure meets the threshold requirement, the PSF value meets the uniformity requirement and the main value parameter meets the preset parameter index.

9. An electronic device comprising a memory and a processor, wherein the memory has stored therein a computer program, the processor being arranged to run the computer program to perform the optical structure optimization method of the lens group of any one of claims 1 to 7.

10. A storage medium having a computer program stored therein, wherein the computer program, when executed by a processor, implements the optical structure optimization method of the lens group of any one of claims 1 to 7.