Differential-based image restoration method and system, storage medium and terminal

Abstract

The invention discloses a difference-based image restoration method and system, a storage medium and a terminal, and the method comprises the steps: carrying out the imaging of a to-be-detected target, obtaining optical system aberrations at different moments through a wavefront detector, and obtaining detection images at different moments through an imaging detector; based on the aberration theory, generating a point spread function (PSF) according to the aberration of the optical system; obtaining point diffusion systems PSF1 and PSF2 corresponding to the target at two different moments and detection images O1 and O2 of corresponding imaging detectors; based on a differential method, performing data processing on the obtained PSF1 and PSF2 and the detection images O1 and O2 to obtain # imgabs0 # and a differential detection image # imgabs1 #, and performing image restoration processing to obtain a restored clear image corresponding to the target; according to the invention, image restoration under the conditions of large aberration and severe noise is realized, and the quality and efficiency of image restoration are improved.

Description

Technical Field

[0001] The present invention relates to the field of computer vision technology, and more particularly to a differential-based image restoration method, system, storage medium, and terminal. Background Art

[0002] Image blur can significantly degrade image quality. Despite extensive research over the past two decades, it remains a research hotspot. Post-processing techniques are often required to improve image quality. However, the effectiveness and accuracy of image restoration algorithms are significantly limited by the influence of optical aberrations and noise.

[0003] Optical aberrations can cause non-uniform changes in images and lead to errors in the point spread function (PSF) and camera detection, which is particularly unfavorable for image classification based on deep learning. In addition, noise can cause random changes in pixel intensity, reduce contrast, and may produce false structures or artifacts, making it difficult for restoration algorithms to accurately distinguish between real image features and variations caused by noise.

[0004] Currently, there are many methods for restoring blurred images, such as Lucy-Richardson (LR) deconvolution, Wiener filtering, ForWaRD, GSM, SV-GSM, SA-DCT, BM3DDEB, and deep learning methods. These studies focus on how to find an appropriate balance between improving image clarity and suppressing noise. Among them, Lucy-Richardson (LR) deconvolution is widely used in astrophysics or microscopy due to its simplicity and high tolerance to noise. However, when the optical aberrations are large, LR deconvolution usually produces obvious ringing artifacts. In addition, an important disadvantage of LR deconvolution is that it will not converge to a meaningful solution because the noise will be amplified after a small number of iterations.

[0005] Therefore, how to eliminate the influence of optical aberrations and noise during the image restoration process and ensure that image restoration can be performed quickly and effectively even in the presence of high noise levels and large aberrations is an urgent problem that technicians in this field need to solve. Summary of the Invention

[0006] In view of this, the present invention provides a differential-based image restoration method, system, storage medium, and terminal to solve some of the technical problems mentioned in the background technology.

[0007] In order to achieve the above object, the present invention adopts the following technical solutions:

[0008] A differential image restoration method includes the following steps:

[0009] S1. Imaging the target to be measured, obtaining the optical system aberration at different times using a wavefront detector, and obtaining detection images at different times using an imaging detector;

[0010] S2. Based on aberration theory, generate a point spread function (PSF) according to the aberration of the optical system;

[0011] S3. Obtain the point spread systems PSF1 and PSF2 corresponding to the target at two different times;

[0012] S4. Obtain detection images O1 and O2 obtained by imaging detectors corresponding to PSF1 and PSF2;

[0013] S5. Based on the differential method, the acquired PSF1 and PSF2 and the detection images O1 and O2 are processed to obtain , and the differential detection image ;

[0014] S6. Based on and Perform image restoration processing to obtain a restored clear image corresponding to the target.

[0015] Preferably, the target to be measured is acquired through an image acquisition device or through a source of image data to be restored.

[0016] Preferably, the content of obtaining the optical system aberration at different times by the wavefront detector is: by changing the ambient temperature of the optical system and the flow velocity of the ambient airflow, random wavefront aberration is generated, and then the wavefront aberration at different times and the corresponding Zernike coefficient of the wavefront aberration are detected by the wavefront sensor.

[0017] Preferably, the specific content of step S2 is: using an atmospheric turbulence model based on Kolmogorov turbulence theory, directly generating the wavefront aberration according to the Zernike aberration coefficient obtained by the wavefront detector, and then calculating the PSF according to the aberration theory.

[0018] Preferably, in step S4, the detection images acquired by the imaging detector at different times are expressed as:

[0019]

[0020] Where I(x,y) is the original image, O1 and O2 are blur images taken under different exposure conditions, and PSF1 and PSF2 are the point spread functions corresponding to O1 and O2, respectively.

[0021] Preferably, in step S5, based on Fourier theory:

[0022]

[0023] Where ΔO=O 2- O1, ΔPSF=PSF 2- PSF1, Δn=n 2- n1, O1, O2 and ΔO are all normalized to between 0 and 1.

[0024] Preferably, the specific content of step S6 is:

[0025] The optical image is reconstructed using D-LR2A, and the (n+1)th image restoration is expressed as:

[0026]

[0027] Where NLR is a nonlinear reconstruction operation, defined as a nonlinear correlation operation between two functions A and B:

[0028]

[0029] in, represents the Fourier transform of A, the function arg{} represents the returned phase angle, and the parameters α and β are adjusted in the range of [-1, 1] to minimize the influence of noise;

[0030] In the initial step, let I 1 =ΔO(x,y) and perform normalization, and iterate repeatedly based on the image restoration formula until the optimal reconstruction is obtained.

[0031] A differential-based image restoration system, based on the differential-based image restoration method, comprises an optical telescope, an optical collimator, an optical beam splitter, a wavefront detector, an optical imaging mirror, an imaging detector, and a data processing and control computer;

[0032] After the optical telescope forms an image of the target, the converged light beam enters the rear-end optical imaging system;

[0033] The light beam is incident on the optical collimator, and the collimated light beam is incident on the optical beam splitter. One beam of light is incident on the wavefront detector for detecting the system aberration of the optical system; the other beam of light is incident on the optical imaging mirror and converged to the imaging detector for data collection of the optical imaging system.

[0034] The data from the wavefront detector and imaging detector are transmitted to the data processing and control computer for differential image restoration.

[0035] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements the image restoration method based on a differential formula.

[0036] A processing terminal includes a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the computer program, the image restoration method based on the differential formula is implemented.

[0037] As can be seen from the above technical solutions, compared with the prior art, the present invention discloses a differential-based image restoration method, system, storage medium, and terminal for high-resolution reconstruction of optical images, aiming to eliminate the influence of optical aberrations and noise, and ensure that image restoration can be performed quickly and effectively even in the presence of high noise levels and large aberrations;

[0038] Unlike traditional blind deconvolution algorithms, this invention uses a wavefront sensor to measure optical aberrations and accurately estimates the PSF of the optical imaging system, avoiding the PSF estimation process required in blind restoration technology. Based on the differential image processing method, the image to be restored is processed to obtain the corresponding clear image, achieving image restoration under conditions of large aberrations and severe noise.

[0039] The present invention not only overcomes the limitations of traditional restoration algorithms in dealing with severe optical aberrations and noise, but also achieves accurate modeling of the optical system by introducing a wavefront sensor, further improving the quality and efficiency of image restoration. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the following briefly introduces the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are merely embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on the provided drawings without paying any creative work.

[0041] Figure 1 A schematic diagram of an image restoration method based on a differential method provided by the present invention;

[0042] Figure 2 A schematic diagram of the wavefront aberration and the corresponding Zernike coefficients of the wavefront detector provided by the present invention;

[0043] Figure 3 A schematic diagram of the image restoration result provided by the present invention;

[0044] Figure 4 A schematic diagram of an image restoration system based on a differential method provided by the present invention;

[0045] Among them, 1-optical telescope, 2-optical collimator, 3-optical beam splitter, 4-wavefront detector, 5-optical imaging mirror, 6-imaging detector, 7-data processing and control computer, 8-target to be measured. DETAILED DESCRIPTION

[0046] The following will clearly and completely describe the technical solutions in the embodiments of the present invention in conjunction with the accompanying drawings. 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 making creative efforts are within the scope of protection of the present invention.

[0047] The embodiment of the present invention discloses an image restoration method based on a differential formula, such as Figure 1 , including the following steps:

[0048] S1. Imaging the target to be measured, obtaining the optical system aberration at different times using a wavefront detector, and obtaining detection images at different times using an imaging detector;

[0049] S2. Based on aberration theory, generate a point spread function (PSF) according to the aberration of the optical system;

[0050] S3. Obtain the point spread systems PSF1 and PSF2 corresponding to the target at two different times;

[0051] S4. Obtain detection images O1 and O2 obtained by imaging detectors corresponding to PSF1 and PSF2;

[0052] S5. Based on the differential method, the acquired PSF1 and PSF2 and the detection images O1 and O2 are processed to obtain , and the differential detection image ;

[0053] S6. Based on and Perform image restoration processing to obtain a restored clear image corresponding to the target.

[0054] In order to further implement the above technical solution, the target to be measured is acquired through an image acquisition device or through a source of image data to be restored.

[0055] In order to further implement the above technical solution, the content of obtaining the optical system aberration at different times through the wavefront detector is as follows: by changing the ambient temperature of the optical system and the flow velocity of the ambient airflow, random wavefront aberrations are generated, and then the wavefront aberrations at different times and the corresponding Zernike coefficients of the wavefront aberrations are detected by the wavefront sensor.

[0056] In this embodiment, the wavefront aberration is composed of a combination of 36 finite Zernike polynomials. The three-dimensional graph of the random wavefront and its corresponding Zernike coefficients is as follows: Figure 2 As shown; the RMS and PV of the wavefront are 0.4573λ and 9.7641λ respectively.

[0057] In order to further implement the above technical solution, the specific content of step S2 is: based on the Kolmogorov turbulence theory through the atmospheric turbulence model, the wavefront aberration is directly generated according to the Zernike aberration coefficient obtained by the wavefront detector, and then the PSF is calculated according to the aberration theory.

[0058] In practical applications, the wavefront detector can be either a Hartmann wavefront detector or an interferometer, as long as it can realize the wavefront aberration detection function; the wavefront detector can be a traditional wavefront detector, but a wavefront detector with higher detection accuracy helps the differential image restoration method to converge faster; the wavefront detector is used for system aberration detection, avoiding the PSF estimation process of the blind restoration image processing algorithm, which helps to improve the system's data calculation efficiency and the temporal resolution of the image to be restored.

[0059] To further implement the above technical solution, it is assumed that the original image I(x, y) is stable over a period of time (e.g., in short exposure mode). Due to the influence of optical aberrations and noise, both the image and PSF captured by the camera will have detection errors. In step S4, the detection images acquired by the imaging detector at different times are expressed as:

[0060]

[0061] Where I(x,y) is the original image, O1 and O2 are blur images taken under different exposure conditions, and PSF1 and PSF2 are the point spread functions corresponding to O1 and O2, respectively.

[0062] In an embodiment, the content of the point spread function PSF generated by the data processing and control computer is: the atmospheric turbulence model is based on the Kolmogorov turbulence theory, the Zernike aberration coefficient obtained by the wavefront detector is directly generated by the program into the wavefront aberration, and then the PSF is calculated according to the aberration theory.

[0063] In order to further implement the above technical solution, in step S5, based on Fourier theory:

[0064]

[0065] Where ΔO=O 2- O1, ΔPSF=PSF 2- PSF1, Δn=n 2- n1, O1, O2 and ΔO are all normalized to between 0 and 1.

[0066] In order to further implement the above technical solution, the specific content of step S6 is:

[0067] The optical image is reconstructed using D-LR2A, and the (n+1)th image restoration is expressed as:

[0068]

[0069] Where NLR is a nonlinear reconstruction operation, defined as a nonlinear correlation operation between two functions A and B:

[0070]

[0071] in, represents the Fourier transform of A, the function arg{} represents the returned phase angle, and the parameters α and β are adjusted in the range of [-1, 1] to minimize the influence of noise;

[0072] In the initial step, let I 1 =ΔO(x,y) and perform normalization, and iterate repeatedly based on the image restoration formula until the optimal reconstruction is obtained.

[0073] In this embodiment, the image restoration method adopts the Lucy-Richardson-Rosen Algorithm (LRRA) algorithm, which combines the Lucy-Richardson algorithm and the Non-linear reconstruction algorithm and has the advantages of being stable and reliable.

[0074] In practical applications, the image restoration method can be either a direct deconvolution image restoration method or an iterative image restoration method. In the embodiment, the peak signal-to-noise ratio (PSNR) and the improved signal-to-noise ratio (ISNR) are used to evaluate the restoration algorithm. The differential LRRA algorithm is used for image processing to help the image to be restored converge faster. The results are shown in FIG. Figure 3 As shown, from Figure 3 It can be seen that it takes about 100 iterations for LRRA to reach stable convergence, while the differential LRRA algorithm only needs about 30 iterations to reach stable convergence during the image restoration process. Moreover, its restoration effect is better than that of the LRRA restoration structure, and its imaging performance is close to the diffraction limit.

[0075] A differential-based image restoration system, based on a differential-based image restoration method, such as Figure 4 , including an optical telescope 1, an optical collimator 2, an optical beam splitter 3, a wavefront detector 4, an optical imaging mirror 5, an imaging detector 6 and a data processing and control computer 7;

[0076] After the optical telescope 1 forms an image of the target 8, the converged light beam enters the rear-end optical imaging system;

[0077] The light beam is incident on the optical collimator 2, and the collimated light beam is incident on the optical beam splitter 3. One beam of light is incident on the wavefront detector 4 for detecting the system aberration of the optical system; the other beam of light is incident on the optical imaging mirror 5 and converged to the imaging detector 6 for data collection of the optical imaging system;

[0078] The data from the wavefront detector and the imaging detector are transmitted to the data processing and control computer 7 for differential image restoration processing.

[0079] In practical applications, an imaging detector refers to a system that can image the detection target, including imaging devices and photoelectric detection devices; a data processing and control computer refers to a system that can process the data after imaging a local area of ​​the solar surface, including corresponding data acquisition devices; optical collimators and optical imaging mirrors can be either transmissive or reflective.

[0080] A computer-readable storage medium stores a computer program, which implements a differential-based image restoration method when executed by a processor.

[0081] A processing terminal comprises a memory and a processor. The memory stores a computer program that can be run on the processor. When the processor executes the computer program, a differential-based image restoration method is implemented.

[0082] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on the differences from other embodiments. Reference can be made to the common and similar parts between the various embodiments. For the devices disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple, and the relevant parts can be referred to the method description.

[0083] The above description of the disclosed embodiments is intended to enable one skilled in the art to implement or use the present invention. Various modifications to these embodiments will be readily apparent to one skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the present invention. Therefore, the present invention is not limited to the embodiments shown herein but is intended to conform to the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A differential-based image restoration method, characterized in that: The following steps are involved: S1. Imaging the target to be measured, obtaining the optical system aberration at different times using a wavefront detector, and obtaining detection images at different times using an imaging detector; S2. Based on aberration theory, generate a point spread function (PSF) according to the aberration of the optical system; S3. Obtain the point spread systems PSF1 and PSF2 corresponding to the target at two different times; S4. Obtain detection images O1 and O2 obtained by imaging detectors corresponding to PSF1 and PSF2; S5. Based on the differential method, the acquired PSF1 and PSF2 and the detection images O1 and O2 are processed to obtain , and the differential detection image ; S6. Based on and Perform image restoration processing to obtain a restored clear image corresponding to the target.

2. The image restoration method based on differential equation according to claim 1, characterized in that: The target to be measured is acquired through an image acquisition device or through a source of image data to be restored.

3. The image restoration method based on differential equation according to claim 1, characterized in that: The content of the optical system aberration at different times obtained by the wavefront detector is as follows: by changing the ambient temperature of the optical system and the flow velocity of the ambient airflow, random wavefront aberrations are generated, and then the wavefront aberrations at different times and the corresponding Zernike coefficients of the wavefront aberrations are detected by the wavefront sensor.

4. The image restoration method based on differential equation according to claim 1, characterized in that: The specific content of step S2 is: using the atmospheric turbulence model based on the Kolmogorov turbulence theory, directly generating the wavefront aberration according to the Zernike aberration coefficient obtained by the wavefront detector, and then calculating the PSF according to the aberration theory.

5. The image restoration method based on differential equation according to claim 1, characterized in that: In step S4, the detection image acquired by the imaging detector is expressed as: Where I(x,y) is the original image, O1 and O2 are blur images taken under different exposure conditions, and PSF1 and PSF2 are the point spread functions corresponding to O1 and O2, respectively.

6. The image restoration method based on differential equation according to claim 1, characterized in that: In step S5, based on Fourier theory: Where ΔO=O 2- O1, ΔPSF=PSF 2- PSF1, Δn=n 2- n1, O1, O2 and ΔO are all normalized to between 0 and 1.

7. The image restoration method based on differential equation according to claim 1, characterized in that: The specific content of step S6 is: Using D-LR 2 A reconstructs the optical image, and the (n+1)th image restoration is expressed as: Where NLR is a nonlinear reconstruction operation, defined as a nonlinear correlation operation between two functions A and B: in, represents the Fourier transform of A, the function arg{} represents the returned phase angle, and the parameters α and β are adjusted in the range of [-1, 1] to minimize the influence of noise; In the initial step, let I 1 =ΔO(x,y) and perform normalization, and iterate repeatedly based on the image restoration formula until the optimal reconstruction is obtained.

8. A differential-based image restoration system, characterized in that: A differential-based image restoration method according to any one of claims 1 to 7, comprising an optical telescope, an optical collimator, an optical beam splitter, a wavefront detector, an optical imaging mirror, an imaging detector, and a data processing and control computer; After the optical telescope forms an image of the target, the converged light beam enters the rear-end optical imaging system; The light beam is incident on the optical collimator, and the collimated light beam is incident on the optical beam splitter. One beam of light is incident on the wavefront detector for detecting the system aberration of the optical system; the other beam of light is incident on the optical imaging mirror and converged to the imaging detector for data collection of the optical imaging system. The data from the wavefront detector and imaging detector are transmitted to the data processing and control computer for differential image restoration.

9. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the image restoration method based on the differential formula described in any one of claims 1 to 7 is implemented.

10. A processing terminal comprising a memory and a processor, wherein the memory stores a computer program that can be run on the processor, characterized in that: When the processor executes the computer program, the image restoration method based on the differential formula as described in any one of claims 1 to 7 is implemented.

Images