Non-intrusive wide-spectrum scattering imaging method for complex target
By building a wide spectrum imaging system, using improved phase recovery and deconvolution algorithms, the non-invasive wide spectrum scattering imaging problem of complex targets is solved, and high-precision complex target reconstruction and real-time rapid imaging are achieved, suitable for multiple fields. application.
Patent Information
- Application Number
- CN202410984182.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-22
- Publication Date
- 2024-11-15
AI Technical Summary
The non-invasive wide spectral scattering imaging that cannot achieve complex goals is not possible. The existing methods are only applicable to simple goals. As the complexity of the target increases, the imaging quality gradually decreases, and the images of complex goals cannot be effectively reconstructed.
By building a wide spectrum imaging system through scattering media, using a wide spectrum LED light source, an improved phase recovery algorithm and a Richardson-Lucy deconvolution algorithm, combined with relaxed average alternating reflection and mixed input and output algorithm, iteratively reconstructs the point diffusion of simple targets function and applied to wide speckle of complex targets to achieve reconstruction of complex targets.
It realizes high-precision reconstruction of complex targets, with simple optical paths, simple operation, robust algorithms, and rapid imaging in real time. It is suitable for biomedical, counter-terrorism and traffic safety fields, and realizes incoherent white light imaging without being limited by the light source spectrum width. .
Smart Images

Figure CN118960953A_ABST
Abstract
Description
Technical Field
[0001] The invention belongs to the technical field of computational optics imaging, and in particular relates to a complex target non-invasive wide-spectrum scattering imaging method. Background Art
[0002] When light waves are transmitted through scattering media (such as clouds, fog, biological tissues, etc.) or interact with their surfaces (such as rough walls, frosted glass, etc.), they will be scattered to varying degrees. Traditional optical imaging systems can only receive speckles formed by the interference of scattered light and cannot directly obtain target information hidden behind the scattering medium.
[0003] At present, the commonly used methods for imaging through scattering media include speckle correlation imaging, transmission matrix light field imaging, and wavefront modulation imaging. Among them, speckle correlation imaging technology can realize the imaging of small targets through scattering media by combining the high similarity between target autocorrelation and speckle autocorrelation with phase recovery technology, but its imaging is limited by the spectral width of the illumination light source. As the spectral width expands, the speckle contrast will gradually decrease, the speckle particle size will change, the speckle structure distinguishability will decrease, and the structural similarity between the target autocorrelation and the speckle autocorrelation will gradually disappear as the spectral width increases. At present, there is no direct method to image targets with wide-spectrum illumination. Both the ideas of imaging through scattering media based on transmission matrix and wavefront modulation require high-quality coherent light source illumination, and perform corresponding wavefront encoding on the light field modulation feedback to obtain the light field information of the front and back surfaces of the scattering medium. In addition to the high quality requirements for the light source, it also requires good system stability, high cost, and inconvenient operation. This technology is difficult to directly promote to wide-spectrum applications. In practical applications, high-performance laser light sources are very expensive. In the fields of biomedicine, cultural relic protection, intelligent transportation and non-destructive detection, it is more practical to use easily available and inexpensive incoherent light sources or white light sources to detect and image scenes.
[0004] The above methods have solved the imaging needs through scattering media to a certain extent, but their imaging spectrum width is limited to laser light sources or narrow-spectrum LED light sources, and usually the spectrum width does not exceed 20nm. Wide-spectrum imaging through scattering media is expected to play a more important role in biomedical imaging, intelligent transportation and other fields, and will hopefully realize passive scattering imaging.
[0005] In the existing research reports on non-invasive wide-spectrum scattering imaging methods based on scattering media, in 2022, Peng Xiang's research group at Shenzhen University used improved phase recovery technology to achieve single-frame non-invasive wide-spectrum illumination through scattering medium imaging for sparse targets and simple targets. Specifically, by adding target sparsity constraints and real-value constraints in the iterative phase reconstruction process, and at the same time applying interpretable constraints to the system optical transfer function in the target Fourier spectrum space, non-invasive single-frame wide-spectrum scattering imaging for sparse targets and simple targets can be achieved. However, the wide-spectrum scattering imaging method based on improved phase recovery technology can only image sparse targets and simple targets. As the complexity of the target increases, the imaging quality gradually decreases, and wide-spectrum scattering imaging of complex targets cannot be achieved. In 2022, Li Wei and others from Xidian University improved the iterative alternating projection algorithm by introducing adaptive support domain shrinkage and hybrid phase recovery measurement, and achieved wide-spectrum non-invasive scattering imaging through scattering media for simple targets. Specifically, the improved iterative alternating projection algorithm is composed of a relaxed average alternating reflectance (RAAR) algorithm and a hybrid input-output (HIO) algorithm, and shrink wrapping is introduced to update the support domain estimation, which realizes wide-spectrum non-invasive scattering imaging of simple targets through scattering media. However, the wide-spectrum scattering imaging method based on the improved iterative alternating projection algorithm can only image simple targets, and cannot achieve wide-spectrum scattering imaging of complex targets.
[0006] Although existing technical means can achieve non-invasive wide-spectrum scattering imaging of simple targets (simple numbers or letters), the quality of target reconstruction decreases as the complexity of the target increases. This is because the cutoff frequency of the pure target Fourier spectrum estimated by spectrum purification technology under wide-spectrum illumination continues to shrink toward the center, so the estimated support domain will also become loose, resulting in a decrease in reconstruction quality. Summary of the invention
[0007] In order to solve the above problems existing in the prior art, the present invention provides a complex target non-invasive wide-spectrum scattering imaging method. The technical problem to be solved by the present invention is achieved by the following technical solutions:
[0008] The present invention provides a complex target non-invasive wide-spectrum scattering imaging method, comprising:
[0009] S1: Building a wide-spectrum imaging system through a scattering medium, the wide-spectrum imaging system comprising a wide-spectrum light source, a simple target, a scattering medium and a camera arranged in sequence along the optical axis direction, the simple target being a digital or letter image;
[0010] S2: using the camera to collect the broad spectrum illumination speckle of a simple target;
[0011] S3: Iteratively reconstructing the simple target by using an improved phase recovery algorithm to obtain a reconstruction result of the simple target;
[0012] S4: Deconvolute the reconstruction result of the simple target using the Richardson-Lucy deconvolution algorithm to obtain a wide spectrum point spread function;
[0013] S5: replacing the simple target with a complex target to be reconstructed, and collecting the wide-spectrum illumination speckle of the complex target using a camera;
[0014] S6: Based on the broad spectrum point spread function, a Richardson-Lucy deconvolution algorithm is used to obtain a reconstruction result of the complex target.
[0015] In one embodiment of the present invention, the broadband light source is a broadband LED light source, and the spectrum width range is 380-760 nm.
[0016] In one embodiment of the present invention, the distance between the broadband light source and the simple target is 10-50 cm, the distance between the simple target and the scattering medium is 5-20 cm, and the distance between the scattering medium and the camera is 40-60 cm.
[0017] In one embodiment of the present invention, the camera is a CMOS industrial camera.
[0018] In one embodiment of the present invention, the broad spectrum illumination speckle I of the simple target 1 It is expressed as:
[0019] I 1 =O 1 *PSF broadband
[0020] Among them, O 1 Represents a simple target, PSF broadband represents the broadband point spread function.
[0021] In one embodiment of the present invention, S3 includes:
[0022] S3.1: setting basic parameters of the improved phase recovery algorithm, including the number of iterations N1 of the relaxed average alternating reflection algorithm, the number of selections N2 of the hybrid input-output algorithm, the feedback coefficient β1 of the relaxed average alternating reflection algorithm, and the feedback coefficient β2 of the hybrid input-output algorithm;
[0023] S3.2: Initialize the reconstruction target ρ (0) , estimated support domain D0 and Gaussian kernel standard deviation σ (0) ;
[0024] S3.3: Reconstruct the target ρ using the relaxed average alternating reflection algorithm and the hybrid input-output algorithm (0) For phase recovery, in a single iteration, the relaxed average alternating reflection algorithm is first run N1 times, and the result is used as a new initial value to enter the hybrid input-output algorithm and iterate N2 times to obtain an updated reconstruction target ρ (j+1) , estimated support region D j+1 and the reconstructed Fourier phase F(ρ (j+1) ), where j+1 represents the j+1th iteration;
[0025] S3.4: Determine whether the reconstruction process converges, and determine whether the difference between the Fourier spectrum reconstructed at the j+1th iteration and the Fourier spectrum reconstructed at the jth iteration is less than a preset value m, that is, calculate ||F(ρ (j+1) )-F(ρ (j) )|| 2 If not, repeat step S3.3 until a qualified reconstruction target is obtained, and finally output the reconstruction target ρ, the estimated support domain D and the reconstructed Fourier phase F(ρ), where the reconstruction target ρ is the simple target finally reconstructed.
[0026] In one embodiment of the present invention, the feedback coefficient β1 of the relaxed average alternating reflection algorithm and the feedback coefficient β2 of the hybrid input-output algorithm are both set to 0.8.
[0027] In one embodiment of the present invention, the expression of the broadband point spread function is:
[0028]
[0029] Where F{} represents Fourier transform, F -1 {} represents the inverse Fourier transform, ε is a positive number between 0 and 1, O 1 Represents a simple target for reconstruction.
[0030] In one embodiment of the present invention, the reconstruction result of the complex target is expressed as:
[0031]
[0032] Where F{} represents Fourier transform, F -1 {} represents the inverse Fourier transform, ε is a positive number between 0 and 1, O 2 Represents the complex target of reconstruction, PSF broadband represents the broadband point spread function.
[0033] Compared with the prior art, the present invention has the following beneficial effects:
[0034] 1. The present invention proposes a complex target non-invasive wide-spectrum imaging method, which mainly solves the problem of complex target non-invasive wide-spectrum scattering imaging through scattering media. Existing methods can only achieve wide-spectrum scattering imaging for simple targets. The complex target non-invasive wide-spectrum imaging method proposed by the present invention can solve the problem of low imaging quality of complex targets during non-invasive wide-spectrum scattering imaging, and can reconstruct the image of complex targets blocked by scattering media, with high reconstructed target accuracy, simple optical path, easy operation, robust and flexible algorithm, and high reconstruction accuracy.
[0035] 2. The present invention can realize imaging of complex targets through scattering media. There is no lens involved in the imaging optical path of the present invention, and it can be applied to lensless wavefront perception. No aberration is introduced during the imaging process, so no aberration correction process is required. The imaging optical path of the present invention does not need to introduce reference light, and there is no light field modulation element, such as SLM or DMD. The optical path is simple and easy to operate, and the reconstruction target accuracy is high. It can realize real-time and rapid imaging observation, and is expected to be applied to passive scattering imaging. The imaging method of the present invention is not limited by the spectral width of the light source illumination, and white light illumination imaging can be realized.
[0036] 3. The present invention can be applied to the fields of biomedical imaging, anti-terrorism and traffic safety, and is not physically limited by the imaging spectrum width, and can achieve incoherent white light imaging.
[0037] The present invention will be further described in detail below with reference to the accompanying drawings and embodiments. BRIEF DESCRIPTION OF THE DRAWINGS
[0038] Figure 1 It is a flow chart of a complex target non-invasive wide-spectrum scattering imaging method provided by an embodiment of the present invention;
[0039] Figure 2 is a schematic structural diagram of a wide-spectrum imaging system through a scattering medium provided by an embodiment of the present invention;
[0040] Figure 3 is a flow chart of an improved phase recovery algorithm provided by an embodiment of the present invention;
[0041] Figure 4 is a schematic diagram of a wide-spectrum illumination speckle, a reconstruction result and a wide-spectrum point spread function of a simple target provided by an embodiment of the present invention;
[0042] Figure 5 is a schematic structural diagram of another wide-spectrum imaging system through a scattering medium provided by an embodiment of the present invention;
[0043] Figure 6 It is a schematic diagram of a wide-spectrum illumination speckle of a complex target and a reconstruction result provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0044] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined purpose of the invention, a complex target non-invasive wide-spectrum scattering imaging method proposed by the present invention is described in detail below in conjunction with the accompanying drawings and specific implementation methods.
[0045] The above and other technical contents, features and effects of the present invention are clearly presented in the following detailed description of the specific implementation modes in conjunction with the accompanying drawings. Through the description of the specific implementation modes, the technical means and effects adopted by the present invention to achieve the predetermined purpose can be more deeply and specifically understood. However, the attached drawings are only for reference and explanation purposes and are not used to limit the technical solutions of the present invention.
[0046] It should be noted that, in this article, relational terms such as first and second, etc. are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants are intended to cover non-exclusive inclusion, so that an article or device including a series of elements includes not only those elements, but also other elements that are not explicitly listed. In the absence of more restrictions, the elements defined by the statement "including one..." do not exclude the existence of other identical elements in the article or device including the elements.
[0047] Embodiment 1
[0048] See also Figure 1 , Figure 1 : is a flow chart of a complex target non-invasive wide-spectrum scattering imaging method provided by an embodiment of the present invention, the imaging method comprising:
[0049] S1: Build a wide-spectrum imaging system through scattering media.
[0050] See also Figure 2 , Figure 2 It is a structural schematic diagram of a wide-spectrum imaging system through a scattering medium provided in an embodiment of the present invention. The wide-spectrum imaging system through a scattering medium includes a wide-spectrum light source, a simple target, a scattering medium and a camera arranged in sequence, and the simple target is a digital or letter image. The wide-spectrum light source is a wide-spectrum LED light source with a spectral width range of 380-760nm, which is used to generate wide-spectrum illumination. The simple target is the letter B, and the scattering medium is frosted glass; the camera is an ordinary COMS industrial camera with a pixel of 2000×2000, which is used to receive speckle field images. In other embodiments of the present invention, the scattering medium may be other scattering media that can form or have high contrast speckles, such as clouds, turbid water, etc.
[0051] In this embodiment, the distance between the wide-spectrum light source and the simple target is about 20 centimeters, the distance between the simple target and the scattering medium is about 10 centimeters, and the distance between the scattering medium and the camera is about 50 centimeters.
[0052] S2: Place a simple target O at the target location 1 , using a camera to collect the wide spectrum illumination speckle of a simple target I 1 , see Figure 4 , Figure 4 is a schematic diagram of a wide-spectrum illumination speckle, reconstruction result and wide-spectrum point spread function of a simple target provided by an embodiment of the present invention, wherein: Figure 4 (a) is the wide spectrum illumination speckle of a simple target. The imaging model of this process can be expressed as:
[0053] I 1 =O 1 *PSF broadband (1)
[0054] Among them, O 1 Represents a simple target, PSF broadband represents the broadband point spread function.
[0055] S3: Iteratively reconstructing the simple target using an improved phase recovery algorithm to obtain a reconstruction result of the simple target.
[0056] This embodiment uses an improved phase recovery algorithm to reconstruct hidden target information. However, the inverse problem of reconstructing the target is highly pathological, so it is necessary to design a highly robust phase recovery algorithm to circumvent this problem. Here, this embodiment improves the traditional phase recovery algorithm, and the improved algorithm is composed of a combination of a relaxed averaged alternating reflector (Relaxed Averaged Alternating Reflectors, RAAR) and a hybrid input-output (Hybrid Input-Output, HIO) algorithm. Specifically, the improved phase recovery algorithm includes the following steps:
[0057] S3.1: Set the basic parameters of the improved phase recovery algorithm, including the number of iterations N1 of the RAAR algorithm, the number of selections N2 of the HIO algorithm, the feedback coefficient β1 of the RAAR algorithm, and the feedback coefficient β2 of the HIO algorithm. An appropriate number of iterations can significantly improve the imaging quality, but too many iterations may cause image quality degradation or increase the computational cost. When the number of iterations exceeds 100, the imaging quality begins to gradually deteriorate.
[0058] S3.2: Initialization settings. Initialize the reconstruction target ρ (0) , the estimated support region D 0 and Gaussian kernel standard deviation σ (0)Using a non-centrosymmetric support region estimation can effectively improve the convergence speed of the algorithm and avoid the "twin image" problem. Therefore, the embodiment of the present invention selects 3 times the target area as the selected initial support region.
[0059] S3.3: Update the reconstruction target ρ (j+1) According to the parameters input in steps S3.1 and S3.2, the RAAR algorithm and the HIO algorithm are used to reconstruct the target ρ (0) Phase recovery is performed. In a single iteration, the RAAR algorithm is first run 80 times, and the result is used as the new initial value to enter the HIO algorithm. It is iterated 10 times and the final reconstruction result is obtained. The negative feedback coefficient β in the RAAR algorithm and the HIO algorithm is set to 0.8. At this time, the updated reconstruction target ρ is obtained. (j+1) , the estimated support region D j+1 and the reconstructed Fourier phase F(ρ (j+1) ), where j+1 represents the j+1th iteration.
[0060] S3.4: Determine whether the reconstruction process converges. Determine whether the difference between the Fourier spectrum reconstructed at the j+1th iteration and the Fourier spectrum reconstructed at the jth iteration is less than a preset constant value m, that is, determine whether the difference between the Fourier spectrum reconstructed at the current iteration and the Fourier spectrum reconstructed at the previous iteration is less than a preset constant value m, that is, calculate ||F(ρ (j+1) )-F(ρ (j) )|| 2 ≤m is true, if not, repeat step S3.3 to continue the next iteration until a reconstruction target that meets the conditions is obtained. Finally, the reconstruction target ρ, the estimated support domain D and the reconstructed Fourier phase F(ρ) are output, where the reconstruction target ρ is the simple target obtained by the final reconstruction. The reconstruction result is as follows Figure 4 (b) as shown.
[0061] S4: Deconvolute the reconstruction result of the simple target using the Richardson-Lucy deconvolution algorithm to obtain the wide spectrum point spread function PSF broadband :
[0062]
[0063] Where F{} represents Fourier transform, F -1 {} represents the inverse Fourier transform, ε is a positive number between 0 and 1, O 1 Represents a simple target for reconstruction. The wide spectrum point spread function PSF obtained based on the Richardson-Lucy deconvolution algorithm broadband The results are as follows Figure 4 (c) as shown.
[0064] S5: replacing the simple target with the complex target to be reconstructed, and using a camera to collect the wide-spectrum illumination speckle I of the complex target 2 , the wide spectrum illumination speckle I of the complex target 2 It is expressed as:
[0065] I 2 =O 2 *PSF broadband (3)
[0066] Among them, O 2 Represents a complex target for reconstruction.
[0067] See also Figure 6 , Figure 6 FIG. 1 is a schematic diagram of a wide spectrum illumination speckle of a complex target and a reconstruction result provided by an embodiment of the present invention. 2 Broad spectrum illumination speckle I 2 like Figure 6 (a) shown.
[0068] It should be noted that the complexity of the target can be determined based on the distribution of image texture features. The specific algorithm uses the pixel co-occurrence matrix to extract the texture features of the image to describe the spatial distribution of the pixels in the image and the overall complexity of the image. The pixel co-occurrence matrix is defined as the joint probability distribution of two pixels appearing simultaneously in the image. The pixel co-occurrence matrix can extract five commonly used feature parameters for calculating the complexity of the image, including image energy, image entropy, image contrast, image homogeneity, and image correlation. The normalized value of the image complexity is calculated based on the above five feature parameters. A value less than 0.5 can be regarded as a simple target, and a value greater than 0.5 can be regarded as a complex target.
[0069] S6: Broad spectrum point spread function PSF calculated based on step S4 broadband And Richardson-Lucy deconvolution algorithm to obtain complex target O 2 The reconstruction result:
[0070]
[0071] Among them, O 2 represents the reconstruction result of the complex target, F{} represents the Fourier transform, F -1 {} represents the inverse Fourier transform, and ε is a positive number between 0 and 1. The reconstruction result of the complex target is as follows Figure 6 (b) as shown.
[0072] The present invention proposes a complex target non-invasive wide-spectrum imaging method, which mainly solves the problem of complex target non-invasive wide-spectrum scattering imaging through scattering media. Existing methods can only achieve wide-spectrum scattering imaging for simple targets. The complex target non-invasive wide-spectrum imaging method proposed by the present invention can solve the problem of low imaging quality of complex targets during non-invasive wide-spectrum scattering imaging, and can reconstruct the image of complex targets blocked by scattering media, with high reconstructed target accuracy, simple optical path, easy operation, robust and flexible algorithm, and high reconstruction accuracy.
[0073] The present invention can realize imaging of complex targets through scattering media. There is no lens involved in the imaging optical path of the present invention, and it can be applied to lensless wavefront perception. No aberration is introduced during the imaging process, so no aberration correction process is required. The imaging optical path of the present invention does not need to introduce reference light, and there is no light field modulation element, such as SLM or DMD. The optical path is simple and easy to operate, and the reconstruction target accuracy is high. It can realize real-time rapid imaging observation, and is expected to be applied to passive scattering imaging. The imaging method of the present invention is not limited by the spectral width of the light source illumination, and white light illumination imaging can be realized.
[0074] The present invention can be applied to the fields of biomedical imaging, anti-terrorism and traffic safety, and is not physically limited by the imaging spectrum width, and can realize incoherent white light imaging.
[0075] In the several embodiments provided by the present invention, it should be understood that the apparatus and method disclosed by the present invention can be implemented in other ways. For example, the apparatus embodiments described above are only illustrative, for example, the division of the modules is only a logical function division, and there may be other division methods in actual implementation, for example, multiple modules or components can be combined or integrated into another system, or some features can be ignored or not executed.
[0076] In addition, each functional module in each embodiment of the present invention may be integrated into one processing module, or each module may exist physically separately, or two or more modules may be integrated into one module. The above-mentioned integrated module may be implemented in the form of hardware or in the form of hardware plus software functional modules.
[0077] The above contents are further detailed descriptions of the present invention in combination with specific preferred embodiments, and it cannot be determined that the specific implementation of the present invention is limited to these descriptions. For ordinary technicians in the technical field to which the present invention belongs, several simple deductions or substitutions can be made without departing from the concept of the present invention, which should be regarded as falling within the scope of protection of the present invention.
Claims
1. A non-invasive wide-spectrum scattering imaging method for complex targets, It is characterized in that include: S1: Building a wide-spectrum imaging system through a scattering medium, the wide-spectrum imaging system comprising a wide-spectrum light source, a simple target, a scattering medium and a camera arranged in sequence along the optical axis direction, the simple target being a digital or letter image; S2: using the camera to collect the broad spectrum illumination speckle of a simple target; S3: Iteratively reconstructing the simple target by using an improved phase recovery algorithm to obtain a reconstruction result of the simple target; S4: Deconvolute the reconstruction result of the simple target using the Richardson-Lucy deconvolution algorithm to obtain a wide spectrum point spread function; S5: replacing the simple target with a complex target to be reconstructed, and collecting the wide-spectrum illumination speckle of the complex target using a camera; S6: Based on the broad spectrum point spread function, a Richardson-Lucy deconvolution algorithm is used to obtain a reconstruction result of the complex target.
2. The complex target non-invasive wide spectrum scattering imaging method according to claim 1, It is characterized in that The broad-spectrum light source is a broad-spectrum LED light source, and the spectrum width range is 380-760nm.
3. The complex target non-invasive wide spectrum scattering imaging method according to claim 1, It is characterized in that The distance between the wide-spectrum light source and the simple target is 10 to 50 centimeters, the distance between the simple target and the scattering medium is 5 to 20 centimeters, and the distance between the scattering medium and the camera is 40 to 60 centimeters.
4. The complex target non-invasive wide spectrum scattering imaging method according to claim 1, It is characterized in that The camera is a CMOS industrial camera.
5. The complex target non-invasive wide spectrum scattering imaging method according to claim 1, It is characterized in that Broad spectrum illumination speckle pattern of the simple target I 1 It is expressed as: I 1 =O 1 *PSF broadband Among them, O 1 Represents a simple target, PSF broadband represents the broadband point spread function.
6. The complex target non-invasive wide spectrum scattering imaging method according to claim 1, It is characterized in that The S3 includes: S3.1: setting basic parameters of the improved phase recovery algorithm, including the number of iterations N1 of the relaxed average alternating reflection algorithm, the number of selections N2 of the hybrid input-output algorithm, the feedback coefficient β1 of the relaxed average alternating reflection algorithm, and the feedback coefficient β2 of the hybrid input-output algorithm; S3.2: Initialize the reconstruction target ρ (0) , estimated support region D 0 and Gaussian kernel standard deviation σ (0) ; S3.3: Reconstruct the target ρ using the relaxed average alternating reflection algorithm and the hybrid input-output algorithm (0) For phase recovery, in a single iteration, the relaxed average alternating reflection algorithm is first run N1 times, and the result is used as a new initial value to enter the hybrid input-output algorithm and iterate N2 times to obtain an updated reconstruction target ρ (j+1) , estimated support region D j+1 and the reconstructed Fourier phase F(ρ (j+1) ), where j+1 represents the j+1th iteration; S3.4: Determine whether the reconstruction process converges, and determine whether the difference between the Fourier spectrum reconstructed at the j+1th iteration and the Fourier spectrum reconstructed at the jth iteration is less than a preset value m, that is, calculate ||F(ρ (j+1) )-F(ρ (j) )|| 2 ≤m is true. If not, repeat step S3.3 until a qualified reconstruction target is obtained. Finally, the reconstruction target ρ, the estimated support domain D and the reconstructed Fourier phase F(ρ) are output, where the reconstruction target ρ is the simple target finally reconstructed.
7. The complex target non-invasive wide spectrum scattering imaging method according to claim 6, It is characterized in that The feedback coefficient β1 of the relaxed average alternating reflection algorithm and the feedback coefficient β2 of the hybrid input-output algorithm are both set to 0.
8.
8. The complex target non-invasive wide spectrum scattering imaging method according to claim 1, It is characterized in that The expression of the broadband point spread function is: Where F{} represents Fourier transform, F -1 {} represents the inverse Fourier transform, ε is a positive number between 0 and 1, O 1 Represents a simple target for reconstruction.
9. The complex target non-invasive wide spectrum scattering imaging method according to any one of claims 1 to 8, It is characterized in that The reconstruction result of the complex target is expressed as: Where F{} represents Fourier transform, F -1 {} represents the inverse Fourier transform, ε is a positive number between 0 and 1, O 2 Represents the complex target of reconstruction, PSF broadband represents the broadband point spread function.
Citation Information
Cited By
Method for carrying out single shooting imaging on target in thick scattering medium by using near-infrared light
CN119715455A
