Visible light image restoration method and device for ground-based large aperture telescope and telescope

Abstract

The application is suitable for the field of astronomical observation imaging technology, and provides a visible light image recovery method and device of a ground-based large-aperture telescope and the telescope, the method comprising: extracting a short-wave infrared band image and a visible light image from a light beam of the telescope for a target image respectively; performing point spread function calculation on the short-wave infrared band image to obtain an infrared point spread function; performing scale transformation on the infrared point spread function to obtain a visible light point spread function; and performing compensation recovery on the visible light image by using the visible light point spread function to obtain a high-resolution visible light image. Since the short-wave infrared band is less affected by atmospheric disturbance, the finally obtained visible light image has high resolution and high resolution, and the image recovery effect is effectively improved, and the phenomenon that the visible light band is difficult to completely describe or overfitting occurs is solved.

Description

Technical Field

[0001] This invention belongs to the field of astronomical observation imaging technology, and particularly relates to a method, device and telescope for visible light image restoration of a ground-based large-aperture telescope. Background Technology

[0002] When imaging a target using a ground-based large-aperture telescope, atmospheric disturbances distort the wavefront phase of the target beam, degrading image quality and preventing the telescope system from reaching its imaging diffraction limit. Current classical methods for overcoming atmospheric disturbances include adaptive optics and image restoration techniques.

[0003] Adaptive optics first uses a wavefront detector to measure wavefront distortion, and then uses wavefront correctors such as deformable mirrors to correct the distorted wavefront. Its main drawback is that there is at least a one-frame time difference between the wavefront detector and the wavefront corrector. This means the wavefront corrector can only compensate using the wavefront distortion measured by the wavefront detector in the previous frame. However, atmospheric disturbances change rapidly, and the point spread function corresponding to the two frames will also change accordingly. Therefore, this method cannot achieve complete correction of wavefront distortion.

[0004] Image restoration techniques utilize the target image generated by telescope imaging and prior knowledge of the imaging process to estimate and restore the target image, thereby improving image resolution. Its main drawback is that visible light is significantly affected by atmospheric disturbances, making target wavefront distortion more complex in the visible light band. Existing models struggle to fully describe this distortion or may exhibit overfitting, greatly reducing the effectiveness of image restoration. Summary of the Invention

[0005] The purpose of this invention is to provide a method, apparatus, and telescope for visible light image restoration of a ground-based large-aperture telescope, aiming to solve the technical problem of poor image restoration effect of ground-based large-aperture telescopes in the prior art.

[0006] In a first aspect, the present invention provides a method for restoring visible light images from a ground-based large-aperture telescope, the method comprising:

[0007] Short-wave infrared and visible light images are extracted from the beam of light directed at the target image by the telescope.

[0008] The point spread function is calculated on the shortwave infrared band image to obtain the infrared point spread function;

[0009] The infrared point spread function is scaled to obtain the visible light point spread function;

[0010] The visible light image is compensated and restored using the visible light point spread function to obtain a high-resolution visible light image.

[0011] Optionally, the step of extracting shortwave infrared and visible light images from the beam of light directed at the target image by the telescope includes:

[0012] A dichroic mirror is used to separate the target image of the telescope into two bands: visible light and short-wave infrared light.

[0013] Images were acquired using a visible light imaging system and a short-wave infrared imaging system, respectively, to obtain short-wave infrared images and visible light images.

[0014] Optionally, the wavelength of the light in the shortwave infrared band image is 0.9µm to 1.7µm.

[0015] Optionally, in the step of calculating the point spread function of the shortwave infrared band image to obtain the infrared point spread function, the PCID (Physically-Constrained Iterative Deconvolution) method is used to calculate the point spread function of the shortwave infrared band image.

[0016] Optionally, the step of scaling the infrared point spread function to obtain the visible light point spread function includes:

[0017] Obtain the optical parameters of the telescope;

[0018] The scale difference between short-wave infrared images and visible light images is determined by the optical parameters.

[0019] The infrared point spread function is scaled based on the scale difference to obtain the visible light point spread function.

[0020] Secondly, the present invention also provides a visible light image restoration device for a ground-based large-aperture telescope, comprising:

[0021] The image separation module is used to extract short-wave infrared and visible light images from the target image of the telescope, respectively;

[0022] The point spread function calculation module is used to calculate the point spread function of the shortwave infrared band image to obtain the infrared point spread function.

[0023] The scaling module is used to perform scaling on the infrared point spread function to obtain the visible light point spread function.

[0024] The compensation and restoration module is used to compensate and restore the visible light image using the visible light point spread function to obtain a high-resolution visible light image.

[0025] Thirdly, the present invention also provides a ground-based large-aperture telescope, comprising:

[0026] Processor; and

[0027] The memory is communicatively connected to the processor; wherein,

[0028] The memory stores readable instructions that, when executed by the processor, implement the method described in the first aspect.

[0029] Fourthly, the present invention provides a computer-readable storage medium having a computer program stored thereon, which, when executed, implements the method of the first aspect.

[0030] The present invention provides a method and apparatus for visible light image restoration using a ground-based large-aperture telescope. In this telescope, a short-wave infrared image and a visible light image are extracted from a beam targeting a target image. The point spread function (PSF) of the short-wave infrared image is then calculated to obtain the infrared PSF. After scaling the infrared PSF, the visible light PSF is obtained. Finally, the visible light PSF is used to compensate and restore the visible light image. Because the short-wave infrared band is less affected by atmospheric disturbances, the resulting visible light image has high resolution, effectively improving the image restoration effect and solving the problems of incomplete description or overfitting in the visible light band. Attached Figure Description

[0031] Figure 1 This is a flowchart illustrating the implementation of a visible light image restoration method for a ground-based large-aperture telescope, as shown in Embodiment 1.

[0032] Figure 2 This is an optical path diagram of a telescope imaging system in the visible light image restoration method for a ground-based large-aperture telescope as shown in Embodiment 1.

[0033] Figure 3 This is a block diagram of a visible light image restoration device for a ground-based large-aperture telescope, as shown in Embodiment 2. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0035] The specific implementation of the present invention will be described in detail below with reference to specific embodiments:

[0036] Example 1:

[0037] Figure 1 This is a flowchart illustrating the visible light image restoration method for a ground-based large-aperture telescope as shown in Embodiment 1. The visible light image restoration method for a ground-based large-aperture telescope shown in Embodiment 1 is applicable to ground-based large-aperture telescopes, which are equipped with a processor to achieve high-resolution restoration of the target image, effectively improving the imaging effect of the target image.

[0038] Step S110: Extract the shortwave infrared image and the visible light image from the beam of light directed by the telescope at the target image.

[0039] Step S120: Calculate the point spread function of the shortwave infrared band image to obtain the infrared point spread function.

[0040] Step S130: Perform a scaling transformation on the infrared point spread function to obtain the visible light point spread function.

[0041] Step S140: The visible light image is compensated and restored using the visible light point spread function to obtain a high-resolution visible light image.

[0042] Because visible light is significantly affected by atmospheric disturbances, wavefront distortion of target images is more complex in the visible light band. Through the steps described above, a ground-based large-aperture telescope extracts short-wave infrared and visible light images from the beam targeting the target image. Then, it calculates the point spread function (PSF) of the short-wave infrared image to obtain the infrared PSF; after scaling the infrared PSF, it obtains the visible light PSF; finally, it uses the visible light PSF to compensate for and restore the visible light image. Since the short-wave infrared band is less affected by atmospheric disturbances, the resulting visible light image has high resolution, effectively improving image restoration and resolving issues such as the inability to fully describe the visible light band or the occurrence of overfitting.

[0043] Specifically, when extracting short-wave infrared and visible light images from the beam of light directed at the target image by the telescope, a dichroic mirror is used to divide the target image of the telescope into two bands: visible light and short-wave infrared light. Images are then acquired using a visible light imaging system and a short-wave infrared imaging system, respectively, to obtain short-wave infrared and visible light images.

[0044] For example, Figure 2This is an optical path diagram of a telescope imaging system according to this embodiment. Wherein, 1 is the first focal point of the telescope; 2 is a collimating convex lens that converts the light beam into parallel light; 3 separates the light beam into visible light and short-wave infrared bands; the visible light band passes through 4, a visible light imaging lens group, which converges the light beam and images it in the visible light band by 5, a visible light camera; the short-wave infrared band passes through 6, a short-wave infrared imaging lens group, which converges the light beam and images it in the short-wave infrared band by 7, a short-wave infrared camera.

[0045] Optionally, this invention is applicable to telescopes with an aperture of 1 meter or more, primarily addressing the problem of decreased image quality caused by atmospheric disturbances in telescopes with an aperture of 1 meter or more. For example, in a 2-meter aperture telescope, the visible light band is 300nm to 700nm, and the short-wave infrared band is 0.9µm to 1.7µm. It should be noted that the band range can be appropriately adjusted as needed in practice.

[0046] Furthermore, when calculating the point spread function of shortwave infrared band images, the PCID (Physically-Constrained Iterative Deconvolution) method can be used, or other methods can be used. The specific point spread function calculation methods will not be described in detail here.

[0047] The concept of PCID (Programmatic Image Restoration) emerged in the blind image restoration problem in the 1990s. Its basic idea is to use the constraints imposed by known physical phenomena on the values ​​of various variables in the mathematical expression as constraints on the inverse problem of image restoration. Based on this, the objective function is rewritten as a constrained objective function, and the solution process is changed to constrained optimization, thereby improving the ill-conditioned nature of the problem and obtaining more accurate results.

[0048] Initially, there were three main physical constraints in physical constraint deconvolution:

[0049] The point spread function has a value greater than or equal to 0 at all points;

[0050] The integral of the point spread function is 1;

[0051] The bandwidth of the point spread function is limited to within the diffraction limit of current optical systems.

[0052] Later, the theory of wave optical diffraction propagation was introduced into PCID. Since the point spread function is a function of the wavefront, and the wavefront can be linearly represented by the Zenke polynomial, the point spread function is thus expressed as a function of the Zenke coefficients. The advantages of doing so are: the three constraints of the original physical constraint deconvolution are naturally satisfied; the scale of the unknowns in the original search for the point spread function was 10,000 dimensions or even higher, but after being linearly represented by the Zenke coefficients, only a few hundred parameters need to be searched, thus reducing the dimensionality of the problem. In fact, not only are the number of variables to be searched reduced, but the orthogonality between the variables is also strengthened. This not only improves the convergence speed of the algorithm, but also improves the accuracy of the algorithm.

[0053] The objective evaluation function is in the following form:

[0054]

[0055] Where Wk(u) represents the weight of frequency u in the k-th frame, which is generally generated by the MTF of the optical system when there is no atmospheric disturbance. Dk(u) is the value of frequency u after the Fourier transform of the acquired k-th frame image. Sk is the OTF corresponding to the k-th frame image, F is the expected spectrum of the desired ideal shortwave infrared image, and γ1 and γ2 are regularization coefficients, which are usually given by experience.

[0056] Using the above formula as the target evaluation function, and substituting it into a search algorithm (such as the BFGS algorithm), F and S are finally obtained. Inverse Fourier transforms are then performed on F and S respectively to obtain the final recovered image and point spread function.

[0057] Because the optical parameters of visible optical systems and short-wave infrared optical systems differ (mainly referring to F#, camera pixel size, camera resolution, etc.), there are inevitably scale differences between short-wave infrared images and visible light images (i.e., the size of the target on the image may be different, resulting in different spatial physical meanings of the target represented by the image). Therefore, it is necessary to perform scale transformation on the two images in advance to convert them to a unified scale, making it possible to replace the point spread function.

[0058] Since the scale difference between the two images can be directly obtained from the parameters of the optical system, the scale difference becomes a fixed value after the optical system is manufactured.

[0059] Therefore, when scaling the infrared point spread function, the optical parameters of the telescope are first obtained; then the scale difference between the short-wave infrared image and the visible light image is determined by the optical parameters; finally, the infrared point spread function is scaled according to the scale difference to obtain the visible light point spread function.

[0060] Finally, the visible light image is reconstructed using the generated visible light image point spread function. The visible light image is compensated and restored using the visible light point spread function; that is, given the degraded image of the target and the point spread function, the original image of the target is derived. The principle formula is:

[0061] D = FS

[0062] Due to issues such as noise and zero points, F cannot be calculated directly using division given D and S.

[0063] It should be noted that when using the visible light point spread function to compensate and restore visible light images, Wiener filtering multi-frame method can be used, or other methods can be used. The specific point spread function calculation methods will not be described one by one here.

[0064] In the multi-frame method of Wiener filtering, the specific calculation formula is as follows:

[0065]

[0066] That is, the spectrum of the degraded image is equal to the product of the target ideal spectrum and the spectrum of the point spread function.

[0067] Example 2:

[0068] Figure 3 This is a block diagram of a visible light image restoration device for a ground-based large-aperture telescope, as shown in Embodiment 2. This device can perform all or part of the steps of any of the visible light image restoration methods for ground-based large-aperture telescopes described above. The device includes:

[0069] Image separation module 10 is used to extract short-wave infrared images and visible light images from the target image of the telescope, respectively;

[0070] The point spread function calculation module 20 is used to calculate the point spread function of shortwave infrared band images to obtain the infrared point spread function.

[0071] The scaling module 30 is used to perform scaling transformation on the infrared point spread function to obtain the visible light point spread function.

[0072] The compensation and restoration module 40 is used to compensate and restore the visible light image using the visible light point spread function to obtain a high-resolution visible light image.

[0073] Example 3:

[0074] Embodiment 3 of the present invention provides a ground-based large-aperture telescope, which can perform all or part of the steps of any of the visible light image restoration methods for ground-based large-aperture telescopes described above. The telescope includes:

[0075] Processor; and

[0076] The memory that is communicatively connected to the processor; wherein,

[0077] The memory stores instructions that can be executed by the at least one processor to enable the at least one processor to perform the method as described in any of the exemplary embodiments above, which will not be elaborated here.

[0078] In this embodiment, a storage medium is also provided. This storage medium is a computer-readable storage medium, such as a temporary or non-temporary computer-readable storage medium that includes instructions. The storage medium may include, for example, a memory for instructions that can be executed by a processor of a server system to complete the aforementioned visible light image restoration method for a ground-based large-aperture telescope.

[0079] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for restoring visible light images from a ground-based large-aperture telescope, characterized in that, The method includes: Short-wave infrared and visible light images are extracted from the beam of light directed at the target image by the telescope. The point spread function is calculated on the shortwave infrared band image to obtain the infrared point spread function; The infrared point spread function is scaled to obtain the visible light point spread function; The visible light image is compensated and restored using the visible light point spread function to obtain a high-resolution visible light image.

2. The method as described in claim 1, characterized in that, The steps of extracting short-wave infrared and visible light images from the beam of light directed at the target image by the telescope include: A dichroic mirror is used to separate the target image of the telescope into two bands: visible light and short-wave infrared light. Images were acquired using a visible light imaging system and a short-wave infrared imaging system, respectively, to obtain short-wave infrared images and visible light images.

3. The method as described in claim 1, characterized in that, The wavelength of light in the shortwave infrared band image is 0.9um to 1.7um.

4. The method as described in claim 1, characterized in that, In the step of calculating the point spread function of the shortwave infrared band image to obtain the infrared point spread function, the PCID method is used to calculate the point spread function of the shortwave infrared band image.

5. The method as described in claim 1, characterized in that, The step of scaling the infrared point spread function to obtain the visible light point spread function includes: Obtain the optical parameters of the telescope; The scale difference between short-wave infrared images and visible light images is determined by the optical parameters. The infrared point spread function is scaled based on the scale difference to obtain the visible light point spread function.

6. A visible light image restoration device for a ground-based large-aperture telescope, characterized in that, The device includes: The image separation module is used to extract short-wave infrared and visible light images from the target image of the telescope, respectively; The point spread function calculation module is used to calculate the point spread function of the shortwave infrared band image to obtain the infrared point spread function. The scaling module is used to perform scaling on the infrared point spread function to obtain the visible light point spread function. The compensation and restoration module is used to compensate and restore the visible light image using the visible light point spread function to obtain a high-resolution visible light image.

7. A ground-based large-aperture telescope, characterized in that, The telescope includes: Processor; and The memory is communicatively connected to the processor; wherein, The memory stores readable instructions that, when executed by the processor, implement the method as described in any one of claims 1-5.

8. A computer-readable storage medium having a computer program stored thereon, the computer program, when executed, implementing the method as described in any one of claims 1-5.

Images